|
<< Click to Display Table of Contents >> Navigation: MultiFocusing > EG Fold 3D |
EG Fold 3D is a preparation and analysis module for the Enhanced Gather (EG) and Multi-Focusing (MF) data enhancement workflow in 3D surveys. It computes the actual fold of raw seismic data within each bin for a user-defined series of offset classes, then optionally calculates the Fresnel Zone (FZ) scaling factor required to reach a desired MF fold target. The results are displayed as colour-coded fold maps and offset panels, giving the processor a clear picture of acquisition coverage before running the full 3D enhancement.
For each bin in the calculation area the module counts how many raw traces fall within the configured CMP and CS-CR aperture windows for every offset class. These counts are collected into per-offset fold matrices and a combined total fold map, all interpolated onto a regular spatial grid. If FZ optimisation is enabled (either at execution time via Calculate FZ on execute or interactively via the Calculate fz matrix verb), the module iterates over FZ factor values from the minimum to the maximum in the specified step to find the smallest factor that achieves the desired MF fold. The resulting per-bin FZ factor is saved to the output matrix and can also be exported as an FZ picking item for use downstream.
Use this module before running any 3D MF-based data enhancement (ZO-MF or CO-MF) to verify that sufficient raw-data coverage exists and to precompute the spatially-varying FZ factor map that controls the enhancement aperture.
When set to false (default), trace header geometry is supplied through the in-memory Input trace headers (of seismic data) connector. When set to true, the module reads trace headers directly from disk via the Input traces data handle connector; this mode is recommended for very large 3D surveys where loading all headers into memory is impractical.
The in-memory trace header index (GTraceVectorItem) of the raw seismic dataset. This connector is active when Use input headers from disk is false. It provides source, receiver, and CMP coordinates needed to evaluate the fold within each aperture.
A handle to the seismic data file on disk (GReadWriteTraceVectorItem). This connector is active when Use input headers from disk is true. Headers are read in bulk from disk; set the Trace handler read bulk size parameter to control memory usage.
The trace header index (GTraceVectorItem) of the output (virtual or enhanced) geometry — the set of CMP bins for which fold is to be computed. This defines the target grid; it is typically the geometry of the enhanced gather dataset that will be produced downstream.
The 3D bin grid definition (GBinGridItem) of the survey. The module uses this to construct the spatial output matrices and to define the grid orientation, extent, and inline/crossline numbering. A valid fold map must exist inside the bin grid — if the fold map is missing or invalid, re-create the bin grid before running this module.
Number of trace headers to read from disk in a single I/O operation when Use input headers from disk is true. Default: 200,000. Increase this value on machines with large RAM to improve I/O throughput; decrease it if memory is limited.
When true, the FZ matrix optimisation step is performed automatically at the end of the main execution pass, immediately after the base fold maps have been built. Default: false. Set to true if you want a fully automated run. If you prefer to first inspect the base fold maps and only then trigger FZ computation interactively, leave this false and use the Calculate fz matrix verb.
Container group holding all parameters that control the Fresnel Zone factor calculation. The FZ factor is a spatial multiplier applied to the standard CMP and CS-CR apertures; a value greater than 1 enlarges the search area so that more raw traces can contribute to each enhanced gather location, thereby increasing the achievable MF fold.
The lower bound of the FZ factor scan range. Default: 1 (no aperture expansion). The scan starts at this value and increments by Step fz factor until the desired fold is reached or Max FZ factor is exceeded. Keeping this at 1 means the module first tests whether the nominal aperture already achieves the target fold.
The upper bound of the FZ factor scan range. Default: 1. If the desired MF fold still cannot be achieved at this maximum factor, the module assigns this maximum value as the FZ factor for that location and moves on. Increase this parameter in areas with sparse acquisition where a larger aperture enlargement is needed. Typical values range from 1.5 to 3.
The increment by which the FZ factor is increased during each iteration of the scan. Default: 0.025. Smaller steps produce a more precise FZ factor map but require more iterations and longer computation time. Larger steps run faster but result in a coarser (step-quantised) FZ factor surface.
Selects which aperture dimension is scaled by the FZ factor during the optimisation scan. Default: cmp and cs-cr. The three options are:
cmp — only the CMP aperture (Max distance to CMP) is multiplied by the FZ factor; the CS-CR aperture remains fixed.
cs-cr — only the source-receiver proximity aperture (Max distance to CS-CR) is scaled; the CMP aperture remains fixed.
cmp and cs-cr — both apertures are scaled simultaneously by the same FZ factor. This is the recommended default for most surveys.
The target number of contributing traces (fold) that the FZ factor optimisation tries to achieve for each bin and offset class. Default: 20. When set to zero, the module calculates the nominal MF fold using the fixed apertures (no FZ scanning). When greater than zero, the module scans the FZ factor range bin-by-bin to find the smallest factor that yields at least this many contributing traces. Higher values generally improve the quality of the MF enhancement but require a larger aperture and more computation time.
Reference (near-surface) velocity used when constructing the FZ picking item. Units: m/s. Default: 2000 m/s. Set this to a reasonable estimate of the shallow velocity in your survey area so that the velocity model embedded in the FZ picking output is physically meaningful.
The maximum two-way travel time used when building the FZ picking item, in seconds. Default: 4.0 s. Set this to cover the full time extent of your seismic data. The picking item is sampled at 4 ms intervals from 0 to this value.
Container group controlling the resolution and spatial interpolation of the output fold and FZ factor maps. These settings affect only the visual display matrices; they do not change the underlying per-bin fold counts.
Grid cell size of the output fold and FZ factor maps in the X (inline) direction, in metres. Default: 50 m. Use a value comparable to the bin spacing for a representative display. Larger values create coarser but faster-to-render maps.
Grid cell size of the output maps in the Y (crossline) direction, in metres. Default: 50 m. Typically set equal to Step x unless the survey has significantly different inline and crossline bin spacings.
Spatial interpolation algorithm used to fill the output map grids from the per-bin fold and FZ factor point clouds. Default: Triangulation. Triangulation (Delaunay) is fast and honours all data points exactly. Abos is a fast distance-weighted method suitable for regular datasets. Kriging provides the smoothest result and is preferred for display purposes in irregularly sampled surveys, but is the slowest option.
Variogram model used by the Kriging interpolator. Default: Exponential. Active only when Interpolation method is set to Kriging. The Exponential model is a safe general-purpose choice. Use Gaussian for data with very smooth spatial variation; use Spherical for data with a clear effective range of spatial correlation.
The spatial correlation range for the Kriging variogram, in metres. Default: 100,000 m. Active only when Interpolation method is Kriging. The large default ensures that the Kriging interpolation uses the global data trend. Reduce this value to confine interpolation to local neighbourhood effects.
The number of nearest neighbouring data points used by the Kriging interpolator for each grid node. Default: 15. Active only when Interpolation method is Kriging. Increasing this number provides smoother results at the cost of computation time.
Container group defining the spatial extent and subsampling of the fold calculation. Use the inline/crossline range limits and optional polygon to focus computation on a subset of the survey, which is useful for quick QC checks or when only part of the 3D grid needs to be analysed.
Optional polygon (GPointVectorItem) that spatially limits the bins included in the fold calculation. Only bins whose XY coordinates fall inside the polygon are processed. When not connected, all bins within the inline/crossline range are used.
The first inline number to include in the calculation. Default: -1 (no limit — use the full inline extent of the bin grid). Set a positive value to restrict processing to a specific inline range.
The last inline number to include in the calculation. Default: -1 (no limit). Must be greater than or equal to First inline number when a limit is applied.
The first crossline number to include in the calculation. Default: -1 (no limit).
The last crossline number to include in the calculation. Default: -1 (no limit).
Determines how the inline/crossline range limits and the polygon mask are combined when selecting bins. Default: Logical AND. With Logical AND, a bin is included only if it satisfies both the inline/crossline range and (if connected) lies inside the polygon. With Logical OR, a bin is included if it satisfies either condition.
Inline subsampling stride: only every N-th inline within the defined range is processed. Default: 1 (every inline). Increase to speed up QC runs over large surveys, at the cost of spatial resolution in the output maps.
Crossline subsampling stride: only every N-th crossline within the defined range is processed. Default: 1 (every crossline). Use together with Step along inLine number for rapid preview calculations.
Container group defining the offset range and class width for per-offset fold analysis. The module divides the offset axis from Offset from to Offset to into equal-width panels of width Offset step, and computes a separate fold map and FZ factor map for each panel. This allows the processor to identify offset ranges where coverage is poor.
Minimum absolute offset to include in the fold analysis, in metres. Default: 0 m. Traces with absolute offset below this value are excluded from all fold counts.
Maximum absolute offset to include in the fold analysis, in metres. Default: 5000 m. Traces with absolute offset above this value are excluded. Set this to the maximum usable offset in your survey.
Width of each offset class panel, in metres. Default: 500 m. For example, with defaults of 0–5000 m at 500 m step, ten offset panels are created (0–500, 500–1000, ..., 4500–5000 m). A finer step produces more panels and a more detailed analysis of offset-dependent fold variation, but increases computation time and memory use.
Container group controlling how raw input traces are spatially selected and filtered when counting fold for each output bin location.
Controls whether the secondary (CS-CR) aperture filter is applied in the source-receiver position domain or in the offset domain. Default: XY aperture.
XY aperture — a raw trace is accepted if the distance between its source and the target source (and its receiver and the target receiver) is within Max distance to CS-CR. This is the standard MF selection criterion.
Offset aperture — a raw trace is accepted if the difference between its absolute offset and the target trace absolute offset is within Max distance to CS-CR. Use this mode for CO-MF workflows where offset matching is more important than source-receiver proximity.
Spatial decimation applied to the raw trace set before the aperture selection. Default: 1 (no decimation). A value of 2 uses every second trace, halving computation time at the cost of fold count accuracy. Use values greater than 1 only for approximate QC runs.
The maximum distance from a raw trace's CMP to the target output bin centre for the trace to be considered a contributor. Units: metres. Default: 150 m. This is the primary (CMP) aperture radius. Set it to approximately half the bin spacing or the planned CMP aperture of the data enhancement module. Valid range: 0–100,000 m.
The secondary aperture radius controlling how closely a raw trace's source and receiver must match the target trace's source and receiver positions (in XY aperture mode) or how closely the offsets must match (in Offset aperture mode). Units: metres. Default: 250 m. This parameter has the strongest influence on the number of traces contributing to each enhanced gather location; it should match the planned CS-CR aperture in the downstream enhancement module. Valid range: 0–100,000 m.
The maximum angular difference (in degrees) between a raw trace's source-receiver azimuth and the target trace's azimuth for the raw trace to be accepted. Default: 360 degrees (all azimuths accepted, effectively disabled). Reduce this value in azimuthally anisotropic surveys where azimuthal fold balance is important for the MF enhancement quality.
When false (default), only the configuration where the raw source matches the target source and the raw receiver matches the target receiver is tested. When true, the swapped configuration (raw source near target receiver and vice versa) is also accepted. Enable this for surveys where source-receiver reciprocity is important or when the effective fold is low with the asymmetric setting.
Selects whether the fold calculation runs on the CPU or on a GPU. The multi-threaded CPU execution is the default and is suitable for all systems. GPU execution can significantly accelerate the aperture-point counting loop on large 3D surveys when a compatible graphics card is available.
Options for distributing the calculation across multiple computing nodes in a cluster environment. Leave at default values for single-machine execution.
Minimum number of bins processed per distributed job chunk. Relevant only when distributed execution is active.
When enabled, restricts the number of parallel threads used on each distributed node. Useful for sharing compute nodes with other jobs.
An optional text label appended to distributed job names to distinguish runs in the cluster job queue.
When enabled, allows specifying a CPU core affinity mask for the worker threads. This is an advanced performance tuning option and is not required for typical use.
The CPU core affinity mask applied when Set custom affinity is enabled. Active only when Set custom affinity is true.
The number of parallel CPU threads used for the fold counting loop. By default, all available cores are used. Reduce this value if you need to leave processing capacity for other applications running simultaneously.
When enabled, this module is bypassed entirely in the workflow and its execution is skipped. Use this to temporarily disable the fold analysis step without removing the module from the processing sequence.
A GVerticalVelocityPickingItem containing the computed per-bin FZ factor values encoded as a function of time, using the V0 and Max time parameters to define the time axis. This item is produced only after the FZ matrix calculation has been performed (either at execution time or via the Calculate fz matrix verb). Connect this output to the FZ picking item input of a downstream 3D data enhancement module to apply spatially varying aperture control.
A GMatrixItem containing the total (summed across all offset classes) fold map interpolated onto the output map grid. Use this map to verify overall raw-data coverage before running the full MF enhancement. Areas with very low fold will produce poor-quality enhanced gathers regardless of the FZ factor used.
A GMatrixItem holding the spatially-varying FZ factor map. Each cell value is the smallest FZ factor (in the range Min FZ factor to Max FZ factor) required to achieve the Desired MF Fold at that location. This map can be connected to the FZ factor input of downstream 3D MF enhancement modules as an alternative to the FZ picking item.
The module provides two map-view display groups, both shown on a 2D location map in the XY plane:
Original — shows the calculation area outline, the per-bin total fold coloured point cloud (Total fold points), the interpolated total fold matrix, and the per-offset fold matrices (Original fold matrices). Use this view to assess the base coverage before FZ optimisation.
New — shows the calculation area outline, the FZ factor coloured point cloud (FZ points), the enhanced fold point cloud (New fold points), the new per-offset fold matrices, and the FZ factor matrices. Use this view after the FZ calculation to verify that the optimised apertures achieve the desired fold target.
Triggers the FZ factor optimisation scan interactively, without re-running the full fold calculation. This verb is available after the module has been executed at least once (so that the base fold maps are populated). Use it to experiment with different values of Min FZ factor, Max FZ factor, and Desired MF Fold without repeating the expensive aperture counting step. After this verb completes, the New display group and the output FZ matrix are updated.
Converts the current FZ factor point cloud into the FZ picking item output without recomputing the FZ matrices. Run this verb after Calculate fz matrix if you want to regenerate the picking item following a parameter change (for example, a different V0 or Max time value) without repeating the full FZ scan.