|
<< Click to Display Table of Contents >> Navigation: Statics > Geomage Surface Consistent Statics Search Distributed |
Note: This module is deprecated and retained for legacy project compatibility only. New projects should use the current Geomage Surface Consistent Statics workflow instead.
This module estimates surface-consistent static shifts using a distributed (multi-node) cross-correlation approach. It operates on CMP gathers — either 2D line or 3D survey geometry — and searches for the time shift that best aligns each trace with its neighbours in a spatial aperture around each CMP bin. The resulting static shifts are decomposed into source, receiver, CMP, and offset-dependent components and written to a binary output file for subsequent application by the Geomage Surface Consistent Statics Solver module.
The search strategy uses trace-to-trace cross-correlation within a spatial super-gather formed by merging neighbouring CMP bins. Cross-correlation functions are cached between adjacent bins to avoid redundant computation. The global surface-consistent solution is refined through repeated Common Offset (CO) iterations, each of which progressively widens the offset and CMP aperture by a user-defined factor. The algorithm supports parallel execution across multiple CPU threads and distributed execution across multiple compute nodes.
The seismic dataset in SEG-Y format from which CMP gathers are read. Connect the SEG-Y data handle item that points to the pre-stacked input data. The module reads gathers by CMP bin as defined by the geometry index.
The gather geometry index that defines how traces are organised into CMP bins, and records the source, receiver, and offset attributes for each trace. This index is used to build the spatial neighbourhoods for the cross-correlation search and to separate the estimated total shifts into surface-consistent components. The index must describe either a 2D CMP line or a 3D CMP grid; any other geometry will produce an error.
The number of cross-correlation search iterations performed within each Common Offset (CO) iteration cycle. At each iteration, individual trace shifts are updated by finding the peak of the stacked cross-correlation function and the global surface-consistent solution is re-solved. Increasing this value allows the alignment to converge more closely but also increases computation time. Default value: 1. For noisy data or large initial static misalignments, values of 3 to 5 are typical.
The number of iterations used by the internal linear solver when decomposing the measured total trace shifts into separate source, receiver, CMP, and offset-dependent static components. A higher count improves the accuracy of the decomposition at the cost of additional processing time. Default value: 100. In most cases the default is sufficient.
The length of the time window (in seconds) used to compute the cross-correlation function between pairs of traces. This window is centred on the search position and must be long enough to capture the dominant signal content but short enough to avoid wrapping cycle-skipping effects. The effective window is always expanded to be at least four times the Maximum Shift length. Default value: 0.064 s. This parameter works together with the Start Time and End Time parameters that restrict the portion of each trace used in the correlation.
The maximum allowable static shift (in seconds) that can be detected for any single trace. The cross-correlation peak search is limited to this range, preventing unrealistic large shifts from being picked. This value should be set to slightly exceed the maximum expected surface-consistent static in the survey. Default value: 0.012 s. Setting this value too large risks cycle-skipping on noisy data.
The beginning of the time gate (in seconds) within each trace that is used for cross-correlation. Samples before this time are zeroed out and do not contribute to the static search. This allows the user to exclude surface noise, direct arrivals, or shallow refractions from the analysis. Default value: -1 (use from the start of the trace). Set a positive value in seconds to restrict the analysis to the reflection zone of interest.
The end of the time gate (in seconds) within each trace that is used for cross-correlation. Samples after this time are zeroed out. This allows deep noise or multiples to be excluded. Default value: -1 (use to the end of the trace). Set a positive value in seconds to define the upper limit of the reflection window.
The half-aperture in the crossline direction, measured in number of CMP bins, used to build the super-gather for each target CMP bin. All bins within this distance in the crossline direction are merged with the target bin before cross-correlation. A larger aperture increases the fold of the super-gather and reduces the effect of noise but may include traces from a geologically different area. Default value: 5 bins. For 2D surveys this parameter controls the CMP range used.
The half-aperture in the inline direction, measured in number of CMP bins, used to build the super-gather. Works together with CrossLine Aperture to define a rectangular spatial neighbourhood around each target bin in 3D surveys. Default value: 5 bins. This parameter is not used in 2D mode.
The number of Common Offset (CO) iteration cycles. In each CO cycle the source-receiver aperture and the CMP aperture are expanded by the CO Iteration Factor, allowing the algorithm to capture progressively longer-wavelength static trends. Increasing this count can improve the recovery of long-wavelength statics but also increases computation time. Default value: 1. Use values of 2 or 3 when a significant long-wavelength static is expected.
The initial source-receiver distance aperture (in metres) used during the first Common Offset iteration cycle to group traces into common-offset sets. Only trace pairs within this source-receiver distance of each other are included in the cross-correlation stacking for that cycle. Default value: 100 000 m (effectively unlimited). Reduce this value to restrict the first CO cycle to near-offset traces.
The initial CMP distance aperture (in metres) used during the first Common Offset iteration cycle to select neighbouring traces for cross-correlation stacking. Only traces whose CMP location falls within this distance of the target CMP are included. Default value: 100 000 m (effectively unlimited). Reduce to limit the stacking neighbourhood to a tighter CMP range for the first cycle.
The multiplicative factor by which both the source-receiver aperture (COSRAperture) and the CMP aperture (COCMPAperture) are expanded at the end of each Common Offset iteration cycle. A factor greater than 1 progressively widens the search area across successive CO cycles, allowing long-wavelength statics to be captured in later cycles. Default value: 2. Only relevant when COIterCount is greater than 1.
An additional scaling factor applied to the CMP aperture at each CO iteration, independent of the COIterFactor. This allows the CMP neighbourhood to grow at a different rate than the offset aperture across successive cycles. Default value: 1 (no additional CMP aperture expansion beyond COIterFactor). Increase to make the CMP aperture grow faster than the offset aperture.
An integer code that selects the variant of the surface-consistent decomposition solver used to separate the total trace shifts into source, receiver, CMP, and offset components. Different values activate different solver strategies. Default value: 2. Consult the Geomage statics documentation for a description of each solver variant.
The path and base name of the binary output file (with extension .dat) into which the computed per-trace static shift data are written. This file is subsequently read by the Geomage Surface Consistent Statics Solver module to compute the final source and receiver static solutions. If the Save Reference Static Model option is enabled, two additional SEG-Y files are written to the same directory: reference.sgy and referenceCMP.sgy, which contain the stacked reference model gathers for quality control.
When enabled, the module writes two additional SEG-Y files alongside the main output: a per-bin stacked reference gather and a corresponding normalised variance gather. These files are useful for quality control — they show the alignment of traces after the static correction has been applied and can reveal residual misalignment or noisy bins. Default value: enabled. Disable to reduce output file size if QC data are not required.
When enabled, the surface-consistent decomposition solver uses only the CMP fold equations, simplifying the system of equations solved for the static components. This can improve stability of the solution in surveys with uneven fold or limited offset diversity, at the cost of suppressing some offset-dependent static information. Default value: enabled. Disable to use the full set of source, receiver, and CMP equations when sufficient fold and offset diversity are available.