|
<< Click to Display Table of Contents >> Navigation: Amplitudes > PZ - Calculate |
PZ - Calculate computes a PZ matching operator — a time-domain filter that reconciles the amplitude and phase relationship between the hydrophone (P) component and the vertical geophone (Z) component recorded in ocean-bottom cable (OBC) or dual-sensor marine acquisition. Because P and Z sensors respond differently to up-going and down-going wavefields, combining them can attenuate receiver-side multiples and improve signal quality, but only if the two components are first matched to each other. This module performs that matching calculation and stores the result as a PZ operator, which is subsequently applied to the data using the companion module PZ - Apply.
The module iterates over every receiver location in the survey. For each receiver it reads the corresponding P and Z gathers, optionally applies a linear moveout correction to align reflections across offsets, passes the gathers through user-defined preprocessing sub-sequences, and then derives the matching operator using the selected calculation method. The resulting operators for all receivers are saved together in an output operator item. Two interactive graphics panels allow you to inspect individual receiver gathers and the computed operator for any location before committing to a full run.
This module supports three configurable preprocessing sub-sequences: P gather preprocessing (applied only to the P component), Z gather preprocessing (applied only to the Z component), and Common preprocessing (applied to both components after their individual preprocessing). These sub-sequences can contain any g-Platform processing modules and allow you to condition the data before the operator is estimated.
The SEG-Y data handle for the hydrophone (pressure) component of the dual-sensor recording. Connect this to the SEG-Y file or dataset that holds the P-component traces. The module reads P gathers receiver by receiver during the operator calculation.
The trace header collection for the P component, containing receiver coordinates, offsets, and other geometry attributes. This is used to group traces into receiver gathers and to build the spatial index of all receiver locations displayed in the graphics panel.
The SEG-Y data handle for the vertical geophone (velocity) component of the dual-sensor recording. Connect this to the SEG-Y file or dataset that holds the Z-component traces. The module reads Z gathers that correspond spatially to each P receiver location.
The trace header collection for the Z component, containing receiver coordinates and geometry attributes for all Z-component traces. The module uses these headers to locate the nearest Z receiver for each P receiver and to assemble Z gathers for operator estimation.
Selects the algorithm used to estimate the PZ matching operator. Default: Stack matching.
Stack matching stacks all traces in the P gather and all traces in the Z gather within the specified offset and time window, producing a single representative trace for each component. It then computes a matched (Wiener) filter that transforms the Z stack into the P stack. This method is robust and effective for most OBC datasets. It uses the Prewhitening level parameter to stabilise the filter inversion.
FK analysis works in the frequency-wavenumber (F-K) domain. For each frequency and wavenumber it uses the water velocity to compute the vertical slowness component and derives the optimal scaling between P and Z. This method exploits the physics of wave propagation in water and can produce more accurate operators when the data has good azimuthal sampling. Selecting FK analysis enables the Water velocity parameter. The operator is computed separately for each shot line within the receiver gather and the results are averaged.
The acoustic velocity of seawater at the recording depth, in m/s. Default: 1500 m/s. This parameter is active only when FK analysis is selected as the calculation method. The water velocity determines the vertical slowness used to scale the Z component in the F-K domain. Use a value appropriate for the actual water temperature and salinity at the survey site; typical deep-water values range from 1480 to 1530 m/s.
When enabled, applies a Linear Moveout (LMO) correction to both the P and Z gathers before computing the matching operator. Default: off. LMO flattens the primary reflection arrivals across offsets, which can improve the quality of the stacked traces used in operator estimation (especially with the Stack matching method). Enabling this option makes the LMO velocity parameter editable.
The velocity used to compute the Linear Moveout correction, in m/s. Default: 1550 m/s. This parameter is editable only when Apply LMO is enabled. Set this value to the average near-surface velocity appropriate for your survey area. A value close to the water velocity (1480–1600 m/s) is typical for OBC data where direct waves and shallow water-bottom reflections dominate the moveout pattern used for operator estimation.
The minimum source-receiver offset (in m) of traces included in the operator calculation. Default: 0 m. Traces with offsets below this value are excluded from the gather stack and operator estimation. Increase this value to exclude very short-offset traces that may be contaminated by direct wave energy or cable noise.
The maximum source-receiver offset (in m) of traces included in the operator calculation. Default: 1500 m. Traces with offsets beyond this value are excluded from the gather stack and operator estimation. Set this to a value that captures the reflections of interest while excluding far-offset traces where moveout residuals or noise may degrade the operator quality.
When enabled, the offset window defined by Min offset and Max offset is applied to the absolute (unsigned) offset value. Default: off. Enable this option when trace headers store offsets with negative signs for one direction of shooting, so that traces from both sides of the receiver are treated symmetrically by the offset filter.
The start of the time window (in seconds) used for the operator estimation. Default: 0 s. Only the portion of each trace between Min time and Max time contributes to the stacked trace and operator calculation. Choose a window that captures the reflections you want the PZ sum to enhance while avoiding muted zones or strongly aliased energy at very early or very late times.
The end of the time window (in seconds) used for the operator estimation. Default: 1 s. Together with Min time, this defines the analysis gate. Select a window that is long enough to include representative primary reflection content but short enough to avoid late-time noise or recording artefacts.
A small stabilisation percentage added to the denominator during matched filter inversion, expressed as a fraction of the signal power. Default: 0.01 %. This parameter applies to the Stack matching method and prevents division by near-zero values at frequencies with very low signal energy. Increase the prewhitening level if the resulting operator appears noisy or unstable. Values between 0.01 and 1 % are typical; very high values will progressively suppress the high-frequency content of the operator.
Controls how SEG-Y data is cached in memory during processing. Tuning these settings can improve performance when reading large P and Z datasets, particularly on systems with limited I/O throughput.
Selects whether processing runs on the CPU or an available GPU. CPU is the default and is suitable for all configurations. GPU execution may accelerate the gather preprocessing sub-sequences if GPU-capable modules are included in those sub-sequences.
Enables distribution of the operator calculation across multiple compute nodes in a cluster environment. When enabled, receiver gathers are split among nodes and processed in parallel, which significantly reduces wall-clock time for large 3D OBC surveys with many receivers.
The number of receiver gathers processed in each distributed work unit. Larger bulk sizes reduce scheduling overhead but may cause uneven load balancing if gathers vary greatly in size.
When enabled, restricts the number of processing threads used on each distributed compute node. Use this to reserve system resources for other concurrent tasks or to avoid memory contention when each node processes very large gathers.
An optional text label appended to the distributed job name. Useful for identifying and tracking multiple concurrent PZ - Calculate runs in a cluster job scheduler.
When enabled, allows explicit control over which CPU cores are used by this module, overriding the default operating system scheduling. Enable this only if you need to isolate the processing to specific cores for performance or resource management reasons.
Specifies the CPU core affinity mask when Set custom affinity is enabled. Consult your system administrator for the correct core numbering on your hardware.
The number of parallel CPU threads used to process receiver gathers. Increasing this value speeds up execution on multi-core workstations. The default uses all available cores. Reduce the thread count if other applications need CPU resources concurrently.
When enabled, the module is bypassed during workflow execution and passes its input data through unchanged. Use this setting to temporarily disable the PZ operator calculation without removing the module from the processing sequence.
The computed PZ matching operator set, containing one time-domain filter for each receiver location in the survey. This item is the primary output of the module and must be connected to the Input operator item of the PZ - Apply module to perform the actual PZ summation on the seismic data.
A read-only counter showing the total number of unique receiver locations found in the P trace headers after execution. This value confirms that all expected receivers were present in the input data.
A read-only counter showing the number of receivers for which a valid matching operator was successfully computed after execution. If this value is significantly lower than the number of input receivers, some receivers may lack sufficient data within the specified offset and time window to compute a stable operator. Check your offset and time window settings and verify that corresponding P and Z gathers are available for all receivers.
PZ - Calculate provides several interactive display panels that update in real time as you adjust parameters or click on a receiver location:
Receiver location map — a 2D map showing all receiver positions as points. Click on any point to select that receiver and update all other panels for that location.
Selected P gather — the raw P gather at the selected receiver. Selected P gather with LMO — the P gather after applying the LMO correction. Selected P gather after processing — the P gather after preprocessing sub-sequences and offset/time windowing.
Selected Z gather — the raw Z gather at the nearest Z receiver. Selected Z gather with LMO — the Z gather after LMO. Selected Z gather after processing — the Z gather after preprocessing and windowing.
Selected Z scaled gather — the Z gather after the computed operator has been applied to it. Compare this with the P gather to evaluate how well the operator matches the two components.
Selected operator — the time-domain PZ matching operator (filter wavelet) computed for the selected receiver. P Stack and Z Stack — the stacked trace used as input to the operator estimation for the P and Z components respectively. Use these panels to verify that the operator has a compact, causal shape and that the stacked traces are free of artefacts.