Volume Sort (Records)

Description Performs various types of sorts and/or windowing on 2D records.
Module(s) Image Gather Processes
Requirements Volume
Related
Works with Gathers

This process performs various types of sorts and/or windowing on 2D records. Can be used to create supergathers by merging gathers from two or more adjacent bins together. 

  1. In the Control Panel, open the Process tab.
  2. At the tab header, click the Add icon and select New Process.
  3. Scroll down and double-click on Volume Sort.
  4. Type a name for the process and click OK.

Configuring Volume Sort

  1. Volume: Select the Volume to sort.
  2. Sort To: Select the method of sorting.

Whole Record Sorts

When sorting pre-migration 2D data to Shot, CMP, or Receiver Station records, the output will be complete records of that type.

There are no parameters in this mode.

Super-Gather Sorts

Super-gather sorts group a number of input records into a single larger output record.

For each dimension being super-gathered, you select an Anchor and Increment that determines the output sequence in that dimension. This sequence must be aligned with the input sequence i.e. each ouput record must fall exactly on an input record.

Each output record consists of the input record at that location, plus the traces from a radius of nearby records. For example: with a radius of 2, each output record consists of the traces from 5 consecutive input records centred on the output location.

There is a single 2D super-gather sort that combines CMP records. There are four types of 3D super-gather sort:

  • Super-Gathers along Inlines - combines crosslines along an inline
  • Super-Gathers along Crosslines - combines inlines along a crossline
  • Super-Gathers from 3D Area - combines a rectangle of records in IL/CL (prevents splitting a workflow job into multiple tasks)
  • Super-Gathers from 3D Cross - combines a cross of records along an IL/CL (interactive use only; not available in a workflow job)

This process only combines records into larger ones, increasing fold. If you wish to stack duplicate traces to increase SNR, follow this with a Partial Stacking process.

Panelled Sorts

2D pre-migration:

  • Common Channel
  • Common CMP-Index
  • Common Receiver-Station-Index

2D post-migration:

  • Common-Offset CMP

2D stack:

  • 2D stack

3D post-migration:

  • Common-Offset Inline
  • Common-Offset Crossline

3D stack:

  • Inline records
  • Crossline records

Between forward and reverse Radon:

  • Common P

When sorting to these domains, building a single complete record would require too much memory to do efficiently.

Instead, Insight creates panels that are a portion of the desired record, with enough overlap between the panels to avoid boundary artifacts. Each panel is processed as a separate record, until sorted to a different domain.

You should normally let Insight select the paneling parameters automatically, but you can override these if necessary for certain advanced workflows.

Note: For Common P, you may sort to common P after Radon Forward, but you must sort back the original P domain before Radon reverse.

Parameters

  • Paneling: Whether to use automatic paneling parameters or override.

Note: Insight is best left to optimise memory usage and run times itself. Furthermore, if you override these values, Insight can no longer guarantee that your processes will not create edge effects at the panel boundaries. Only override these if your workflow requires specific panel sizes and you're absolutely certain that you know what you're doing! Talk with support if you're not sure.

  • Move-up: The number of traces to advance between each record. That is to say, the number of traces (from the centre of the panel) that will be kept when the records are viewed or sorted to another domain.
  • Panel Size: The total number of traces in each record.

The panel size and move-up are important, as they dictate how many records are created, and therefore how much processing occurs.

The move-up controls how many traces from each panel will be kept when viewing the data, or sorting to another domain. A move-up of 1 will keep only the one centre trace from the record.

The difference between the panel size and the moveup is the number of traces that will overlap the neighbouring records (half on one side and half on the other). This overlap is normally chosen automatically based on the window sizes of the processes that will operate on the panel records. A panel size equal to the moveup means that all traces will be kept, i.e. no overlap between records. Larger panel sizes and more overlap between records requires more memory and processing time.

Examples:

In this example, panel size = 7 and move-up = 3. Each 7-trace panel will overlap its neighbours by 2 traces on each side. After processing, the centre 3 traces of each panel will be kept, and written to the output volume.

In this example, panel size = 5 and move-up = 5. There's no overlap between the 5-trace panels, and all 5 traces are written to the output volume.

Headers Used

Obviously the specific headers used will depend on the sort mode, but could include:

  • EP - DUG shot point
  • TRACR - DUG channel
  • CDP - DUG CMP
  • HCS - DUG CMP Index
  • CDPT - DUG receiver station
  • GAPS - DUG receiver station index

Headers Modified

  • NHS - Set to the number of live traces in the record. This header is not set for panelled sorts.