|
<< Click to Display Table of Contents >> Navigation: Velocity > Create depth velocity |
The Create depth velocity module converts a velocity model from its source representation into the internal depth velocity format used by depth-domain processing operations such as depth migration, tomography, and model building. It accepts velocity information from three alternative sources: a velocity gather (which stores interval velocities as a function of depth), a constant velocity value, or a grid velocity model derived from NIP (Normal Incidence Point) tomography.
Use this module whenever downstream depth processing steps require a depth velocity object. When a grid velocity model is used as input, the module samples the model onto a regular depth grid at the bin locations defined by a trace vector, producing both the depth velocity object and an output gather in the velocity display domain. The result can be visualized immediately in the inline velocity gather view provided in the module's output.
The input data items available depend on the selected Input velocity type setting. Only the items relevant to the chosen type are active at any time.
This group contains all settings that control how the depth velocity object is built, including the source type selector and all source-specific input fields and numerical parameters. Parameters within this group are shown or hidden automatically depending on the selected input velocity type.
Selects the source from which the depth velocity model will be built. Three options are available:
Gather — builds the depth velocity directly from an existing velocity gather. This is the most common choice when interval velocities have already been derived and are stored as a seismic gather object (for example, from a previous velocity analysis or conversion step). Only the Gather input field is shown.
Const Velocity — reserved for a future uniform-velocity model. This option is not currently supported and will produce an error if selected. Do not use it in production workflows.
Grid velocity — builds the depth velocity from a spatially varying NIP tomography velocity model (grid). The model is sampled at each bin location defined by the Trace vector, at regular depth steps from zero down to the specified maximum depth. Use this option after a refraction or reflection tomography run to propagate the tomographic velocity field into the depth processing chain. Additional parameters (Max depth, Depth step, Scaler factor) and a Grid velocity input field become active.
Default: Gather
The velocity gather that provides the interval velocity values in depth. This field is active only when Input velocity type is set to Gather. Connect the output of a velocity conversion or velocity analysis module here. The gather must contain valid data — if it is empty the module will report an error and will not produce output.
A single velocity value (in m/s) intended to define a uniform velocity model throughout the depth section. This field is active only when Input velocity type is set to Const Velocity. Default: 1500 m/s. Note: this mode is not yet supported in the current software version.
The spatially distributed NIP (Normal Incidence Point) tomography velocity model to be converted into a depth velocity object. This field is active only when Input velocity type is set to Grid velocity. Connect the output of a tomographic inversion or model import step here. The model may be 2D or 3D; the module handles both automatically. Before sampling, the model is scaled by the Scaler factor value.
The deepest depth level (in metres) at which the grid velocity model will be sampled when building the depth velocity gather. Traces in the output gather will extend from zero down to this depth. Set this value to cover the full depth range of interest for your migration or model building workflow. Increasing this value increases processing time when the grid is large.
Active only when Input velocity type is Grid velocity. Default: 4000 m
The vertical sampling interval (in metres) used when sampling the grid velocity model from zero down to the Max depth. Smaller values produce a finer depth sampling and a more detailed velocity gather, at the cost of larger data volumes and longer processing time. Choose a depth step consistent with the expected vertical velocity variation in your model — for smoothly varying models, a coarser step (e.g. 20–50 m) is usually sufficient.
Active only when Input velocity type is Grid velocity. Default: 10 m
A multiplicative factor applied uniformly to all velocity values in the grid model before the depth velocity is built. Use this to rescale a tomographic model — for example, to convert between unit systems, to apply a global velocity correction, or to test the sensitivity of a migration result to velocity perturbations. A value of 1 leaves the model unchanged. A value of 1.05 increases all velocities by 5%.
Active only when Input velocity type is Grid velocity. Default: 1 (no scaling)
Defines the set of bin locations (trace positions) at which the grid velocity model is sampled to produce the output depth velocity gather. This is typically the geometry trace vector of the seismic dataset being processed. Each trace position in the vector becomes one velocity trace in the output gather, sampled from the grid model at that location using the specified Depth step and Max depth. The Trace vector must be connected and non-empty; if it is missing, the module reports an error.
Active only when Input velocity type is Grid velocity.
Selects whether processing is performed on the CPU or a compatible GPU. For most velocity model conversion tasks the CPU option is sufficient. GPU acceleration may benefit large 3D grid models with many sample points.
Controls whether the module runs on the local machine or is distributed across a cluster of processing nodes. Enable this option to offload the computation to remote nodes when working with very large velocity grids.
The minimum number of traces processed together as a single work unit during distributed execution. Larger values reduce scheduling overhead but may cause uneven load balancing across nodes.
When enabled, restricts the number of CPU threads used on each remote processing node. Use this to prevent the module from monopolising a shared node that is also running other jobs.
An optional text label appended to the distributed job name. Use this to distinguish multiple simultaneous runs of this module in the cluster job queue.
When enabled, allows the user to specify which CPU cores are used for this module via the Affinity setting below. Leave disabled to let the operating system assign cores automatically.
Specifies the CPU core mask used when Set custom affinity is enabled. This is an advanced setting for environments where precise CPU resource allocation is required.
The number of parallel CPU threads used during processing. For the Grid velocity mode, increasing the thread count speeds up the sampling of the velocity model across all trace locations. Leave at the default (use all available cores) unless resource contention is a concern.
When enabled, this module is bypassed and its outputs are left unchanged from the previous run. Use the Skip option to temporarily disable the module during workflow testing without removing it from the processing sequence.
The primary output: a depth velocity object in the g-Platform internal format, ready for use by depth migration, tomographic update, and model building modules. This item is always produced, regardless of the input velocity type selected. Connect this output to the depth velocity input of any downstream depth-domain processing step.
The scaled grid velocity model that was used to produce the depth velocity, output as a NIP velocity grid item. This output is produced only when Input velocity type is set to Grid velocity. It contains the model after the Scaler factor has been applied, so it reflects exactly the velocity field used in the conversion. Use this output to verify the scaled model or to pass it to other modules that accept a grid velocity input.
The velocity model represented as a seismic gather, with one velocity trace per bin location. This output is available for all input velocity types and can be displayed in the inline velocity gather viewer that is built into the module. It is useful for visually checking the velocity field before using it in depth migration — anomalous velocities, gaps, or abrupt lateral variations can be identified directly in this display.