Edit azimuthal velocity item

<< Click to Display Table of Contents >>

Navigation:  Velocity >

Edit azimuthal velocity item

 

Description

The Edit azimuthal velocity item module is a bidirectional converter between an azimuthal velocity model item and its three constituent component gathers. An azimuthal velocity model encodes anisotropic NMO velocity information as three spatially distributed attributes: the isotropic (average) RMS velocity, the azimuth of the fast-velocity symmetry axis, and the ellipticity that quantifies the velocity anisotropy strength.

The module operates in two directions, selected by the Edit mode parameter. In Item to gathers mode, it unpacks an existing azimuthal model into three separate editable gathers, optionally re-interpolating them onto a new spatial geometry. In Gathers to item mode, it assembles three independently prepared gathers back into a single azimuthal velocity model item that can be passed to azimuthal NMO or migration modules.

This module is typically used in workflows that require editing or QC of azimuthal anisotropy parameters, changing the spatial sampling of the model, or constructing a model from scratch by supplying custom attribute gathers.

Input data

Azimuthal Vrms model input

Connect the azimuthal velocity model item that you want to unpack. This item holds the three component gathers (isotropic Vrms, azimuth angle, and ellipticity) in a compact structure. This input is required when Edit mode is set to Item to gathers. The model must contain all three valid component gathers; if any component is missing or corrupt, execution will stop with an error.

Interpolate on input geometry

This group controls whether the extracted gathers should be re-interpolated onto a different spatial grid. When enabled, the module reprojects each sample of all three component gathers from the original model geometry to the locations specified by the input geometry trace headers, using the configured interpolation method.

Interpolate on input geometry

Enable this option to re-interpolate the three extracted attribute gathers onto the spatial positions provided by the Input geometry trace headers item. When disabled (default), the output gathers retain the original spatial sampling of the model. Enable this when you need the azimuthal model on a different bin grid, for example to match the geometry of a seismic dataset for azimuthal NMO application.

Input geometry trace headers

Connect the trace header collection that defines the target output locations. Each trace in this header list provides an X/Y bin coordinate; the module will interpolate all three component gathers at each of those locations. This item becomes active only when Interpolate on input geometry is enabled.

Interpolation parameters

This group configures the spatial interpolation method used when re-projecting the three model attribute gathers onto the input geometry. It becomes active when Interpolate on input geometry is enabled. The same interpolation engine is applied identically to all three components (isotropic Vrms, azimuth, and ellipticity).

Interpolation method { Triangulation, Kriging }

Selects the algorithm used to spatially interpolate the attribute gathers onto the new geometry (applies in Item to gathers mode when re-projection is active). Default: Kriging. Triangulation is a fast Delaunay-based method that performs exact piecewise-linear interpolation and is most appropriate when the source bins are regularly or uniformly distributed. Kriging is a geostatistical approach that provides smoother results and extrapolates more gracefully beyond the convex hull of the source points; it is preferred when bins are irregularly spaced or when some areas have sparse coverage.

Kriging covariance type { Exponential, Spherical, Gaussian }

Selects the variogram model used by the Kriging interpolator (active only when Kriging is selected). Default: Exponential. All three model types approach the sill value asymptotically at large distances, but differ in the rate of spatial de-correlation. Exponential gives the sharpest local variability and is generally the most robust default choice. Spherical reaches the sill at exactly the range distance and is common in geophysical applications. Gaussian produces the smoothest interpolation and is suitable when the attribute field is expected to vary very gradually in space.

Kriging range

The spatial correlation range for Kriging interpolation, in metres. Default: 100,000 m. This value defines the distance beyond which two points are treated as spatially uncorrelated. A large range (as supplied by default) effectively makes Kriging behave like a weighted-average smoother across the entire dataset. Reduce this value if you expect the azimuthal velocity attributes to vary significantly over shorter distances and want the interpolation to honour local trends more tightly.

Kriging number of points

The number of nearest source points used to estimate each output location during Kriging. Default: 20. Increasing this value improves the smoothness and stability of the result but increases computation time. Decreasing it speeds up processing but may introduce local artefacts, particularly in areas with uneven source point density. The minimum allowed value is 1.

Input isotropic azimuth gather

Connect the gather containing the isotropic (azimuth-averaged) RMS velocity values, expressed in m/s, for each trace position and time sample. This is required in Gathers to item mode. The isotropic Vrms forms the baseline velocity component of the azimuthal model. Typically this gather is produced by conventional isotropic velocity analysis or by a previous Item to gathers run followed by editing.

Input azimuth gather

Connect the gather containing the azimuth of the fast-velocity symmetry axis, stored in radians, for each trace position and time sample. This is required in Gathers to item mode. The azimuth attribute encodes the direction of the principal anisotropy (for example, the orientation of aligned fractures or stress). It must match the same spatial coverage and sample interval as the isotropic Vrms and ellipticity gathers.

Input ellipticity gather

Connect the gather containing the ellipticity attribute (dimensionless), which describes the degree of azimuthal variation in NMO velocity. This is required in Gathers to item mode. A value of zero represents isotropy; larger values indicate stronger velocity anisotropy. Together with the isotropic Vrms and the azimuth, ellipticity fully parameterises the elliptic approximation to azimuthally anisotropic NMO.

Parameters

Edit mode { Item to gathers, Gathers to item }

Selects the direction of conversion. Default: Item to gathers.

Item to gathers — reads an existing azimuthal Vrms model item and extracts its three constituent gathers (isotropic Vrms, azimuth angle, and ellipticity) as separate output data items. Optionally re-interpolates onto a new spatial geometry. Use this mode to inspect, QC, or manually edit the individual components of an azimuthal model.

Gathers to item — assembles three input gathers (isotropic Vrms, azimuth angle, and ellipticity) into a new azimuthal velocity model item. Use this mode after editing or creating the component gathers externally to package them into the model structure required by downstream azimuthal processing modules.

Changing this parameter toggles the visibility of inputs and outputs: only items relevant to the selected mode are shown in the interface.

Start time { From gathers, Manual }

Controls how the start time of the assembled azimuthal model is determined (active in Gathers to item mode only). Default: From gathers.

From gathers — the start time is read automatically from the header of the input isotropic azimuth gather. Use this option when the gathers already carry the correct time reference.

Manual — the start time is taken from the Manual start time parameter below. Use this when the gathers have been cropped, re-sampled, or otherwise modified so that their embedded time reference no longer reflects the correct model origin time.

Manual start time

The two-way travel-time at the first sample of the assembled model, in seconds. Default: 0 s. This parameter is active only when Start time is set to Manual. All three component gathers will have their internal time reference overridden to this value before being packed into the model item. Ensure that this matches the datum of the seismic data to which the azimuthal model will be applied.

Interpolation parameters

This group configures the spatial interpolation embedded in the output azimuthal velocity model item (active in Gathers to item mode). These parameters define how the model will interpolate attribute values at arbitrary query positions when it is later sampled by downstream modules such as azimuthal NMO or migration. They do not affect the input gather values themselves.

Interpolation method { Triangulation, Kriging }

Selects the spatial interpolation method that the assembled model will use internally when queried for attribute values at trace positions that do not coincide exactly with an input gather location. Default: Kriging. The choice affects the smoothness of the model response when it is applied during azimuthal NMO correction. Triangulation is faster and honours the data exactly at the input positions. Kriging provides smoother model behaviour between control points and is more robust when coverage is irregular.

Kriging covariance type { Exponential, Spherical, Gaussian }

Selects the variogram model for the model item's internal Kriging interpolator (active only when Kriging is selected). Default: Exponential. The covariance type controls the spatial smoothness of the azimuthal attributes between control points when the model is later queried. Exponential is the most commonly used and provides a good balance between smoothness and local accuracy. Spherical reaches zero spatial correlation at the range distance. Gaussian is the smoothest option and is best when anisotropy parameters are expected to vary gradually across the survey area.

Kriging range

The spatial correlation range for the model item's internal Kriging interpolator, in metres. Default: 100,000 m. This controls how far the influence of each control point extends when the model is queried by downstream modules. A large value (default) produces a smooth, regionally averaged behaviour. Reduce this if the azimuthal anisotropy is expected to change significantly over shorter lateral distances.

Kriging number of points

The number of nearest control points used by the model item's Kriging interpolator when queried by downstream modules. Default: 20. Higher values increase the smoothness and stability of the model response at the cost of slightly slower access times. The minimum allowed value is 1.

Settings

Execute on { CPU, GPU }

Selects the hardware used to run the interpolation calculation. Default: CPU. For most dataset sizes this module is CPU-bound and runs efficiently using multi-threaded execution on standard CPU hardware.

Distributed execution

Enables distributed (cluster) execution across multiple compute nodes. Use this option when processing very large geometry interpolation jobs that exceed the capacity of a single workstation.

Bulk size

The number of traces processed in each parallel work unit when distributed execution is active. Larger values reduce inter-node communication overhead but may increase memory use per node.

Limit number of threads on nodes

When enabled, restricts the number of CPU threads used on each distributed node. This can be useful to leave resources available for other processes running simultaneously on the same node.

Job suffix

An optional text label appended to the distributed job name. Useful for identifying jobs in the cluster queue when multiple instances of this module are running simultaneously.

Set custom affinity

Enables manual specification of CPU core affinity. When enabled, the Affinity parameter becomes available. Leave disabled unless you have a specific reason to pin the process to particular CPU cores.

Affinity

Specifies the CPU core or core set to which the process should be pinned. Active only when Set custom affinity is enabled.

Number of threads

The number of CPU threads used for parallel processing of the spatial interpolation. The default uses all available cores. For the re-projection step in Item to gathers mode, the output trace loop is parallelised across threads, so this setting has the most impact when the output geometry contains a large number of traces.

Skip

When enabled, this module is bypassed entirely and execution passes directly to the next module in the workflow. Use this to temporarily disable the conversion step during testing or workflow development without disconnecting the module.

Output data

Output isotropic azimuth gather

The extracted isotropic (azimuth-averaged) RMS velocity gather, produced in Item to gathers mode. Each trace corresponds to one bin position; each sample is the isotropic Vrms value in m/s at that time. If geometry re-interpolation is enabled, the spatial sampling of this gather matches the input geometry trace headers. Connect this output to a velocity display or an editing module to inspect or modify the isotropic velocity field.

Output azimuth gather

The extracted fast-axis azimuth gather, produced in Item to gathers mode. Values are stored in radians, with the display unit converted to degrees for QC. This gather encodes the direction of the maximum NMO velocity at each bin and time. Connect to an attribute display to map the spatial distribution of the anisotropy symmetry axis across the survey.

Output ellipticity gather

The extracted ellipticity attribute gather, produced in Item to gathers mode. Values are dimensionless; zero indicates isotropy and larger values indicate stronger azimuthal velocity variation. Connect this output to an attribute display or editing module to inspect or adjust the strength of the anisotropy throughout the model volume.

Output azimuthal Vrms model

The assembled azimuthal velocity model item, produced in Gathers to item mode. This item packages the three input gathers together with the configured spatial interpolation scheme into a single self-contained model structure. Connect this output to the azimuthal velocity input of any downstream module that requires azimuthal NMO parameters, such as an azimuthal NMO correction or azimuthal Kirchhoff migration.

Information

Graphics

Custom actions