|
<< Click to Display Table of Contents >> Navigation: Stack > CMP common offset stack |
The CMP Common Offset Stack module produces a set of stacked traces, each representing a specific offset range within a CMP (Common Midpoint) gather. It divides the full offset range into a series of bins defined by a minimum offset, a maximum offset, and an offset increment. For each bin, all input traces whose offset falls within that bin are summed and normalized, yielding a single output trace assigned the center offset of that bin.
This module is useful for compressing prestack data into a smaller set of offset-representative traces while preserving offset-dependent amplitude information (AVO), suppressing random noise through stacking, and providing a quick quality-control view of the gather sorted by offset class. It is commonly applied before AVO attribute extraction, partial-offset stack generation, or as a preprocessing step for migration.
Processing is performed in parallel across offset bins to maximize throughput. The number of output traces equals (Maximum offset - Minimum offset) / Offset increment, rounded to the nearest integer.
The prestack seismic gather to be stacked by offset. This is typically a CMP gather sorted by offset, containing multiple traces that span the desired offset range. Each trace must carry a valid offset header value so that the module can assign it to the correct output offset bin. If the gather is empty or invalid, the module reports an input error and skips processing.
Controls how zero-value samples are treated during the stacking operation. When enabled (default), zero-amplitude samples are excluded from both the sum and the sample count used for normalization. This prevents muted or padded regions from diluting the stack amplitude, which is the correct behavior when input traces have been muted before stacking. When disabled, all traces within the offset bin contribute to the denominator regardless of whether their sample value is zero, which produces a true fold-normalized average. Use the disabled setting only when zeros represent genuine signal rather than muted zones.
Default: Enabled (true).
The starting offset value (in meters) of the first output offset bin. Input traces with offsets below this value are not included in any output bin. For conventional acquisition with positive offsets only, set this to 0 or the near-offset value of your survey. For surveys that record negative offsets (e.g., 2D split-spread geometry), set a negative value to capture near-offset traces on both sides of the source.
Default: -4500 m. Units: meters.
The ending offset value (in meters) of the last output offset bin. Input traces with offsets above this value are not included in any output bin. Set this to match the maximum source-receiver offset recorded in your survey. Together with the Minimum offset and Offset increment, this value determines the total number of output traces: (Maximum offset - Minimum offset) / Offset increment.
Default: 4500 m. Units: meters.
The width of each offset bin (in meters). All input traces whose offset falls within a given bin — from bin start (inclusive) to bin end (inclusive) — are stacked together to produce one output trace. The output trace is assigned an offset equal to the center of the bin. Smaller increment values produce more output traces with narrower offset ranges and better offset resolution, but require more traces per gather to ensure adequate fold in each bin. Larger values produce fewer, broader bins with higher fold per bin and better noise suppression.
A typical choice is to set the increment equal to the receiver interval of the acquisition, so that each output bin corresponds to one nominal offset class. For AVO analysis, use a smaller increment (e.g., 100–250 m) to preserve offset-dependent amplitude variation. For noise suppression or display purposes, a larger increment (e.g., 500–1000 m) is appropriate.
Default: 1000 m. Units: meters.
The stacked output gather containing one trace per offset bin. Each output trace represents the normalized sum of all input traces whose offset falls within the corresponding bin. The offset header of each output trace is set to the center value of its bin. The output gather has the same time sampling and record length as the input gather. Offset bins that contain no input traces (zero fold) produce zero-valued output traces.