Check geometry by first breaks

<< Click to Display Table of Contents >>

Navigation:  Geometry >

Check geometry by first breaks

 

Description

This module is an interactive visual tool for verifying and correcting seismic geometry using automatically picked first-break arrivals. It compares the expected first-break hyperbola (derived from a user-specified velocity) against the auto-picked first-break times in each common-point gather, enabling quick identification of receivers or sources whose recorded positions do not match the true geometry.

For each gather, the module automatically picks first breaks using an energy-based characteristic function, then computes a linear moveout (LMO) reference curve. Gathers where the auto-picked first-break arrival and the nearest-offset trace are inconsistent signal a potential geometry error. The operator can browse gathers interactively using the Start slide, Next, and Prev actions, review a location map of all common points, and inspect the collection view to compare multiple gathers simultaneously. Changed geometry points are recorded in an output table for downstream correction workflows.

Input data

Input DataItem

The main seismic dataset container. Connect the project data item that holds the seismic traces to be inspected for geometry correctness.

Input SEG-Y data handle

The SEG-Y file handle providing access to the seismic trace amplitudes. The module reads individual gathers directly from this handle during interactive browsing.

Input trace headers

Trace header metadata including source and receiver coordinates, offsets, and any grouping keys used to form gathers. The module uses these headers to identify and locate each common-point gather on the map.

Input gather

The gather index structure that defines how traces are grouped into common-point gathers (e.g., common mid-point or common receiver gathers). This determines which traces are displayed together when browsing.

Input stack line

Optional 2D line geometry input used to provide spatial context for the location map display.

Input crooked line

Optional crooked-line geometry input for 2D surveys with non-straight acquisition lines.

Input bin grid

Optional bin grid definition for 3D surveys, used to relate common-point positions to inline/crossline coordinates on the location map.

Input sorted headers

The sorted gather index vector that provides the ordered list of all common-point gathers. This is the primary navigation structure: the module iterates through all non-empty gathers defined here to build the location map and pre-compute first-break picks.

Parameters

First

The primary gather selector value, corresponding to the first grouping key of the input sorted headers (e.g., the first gather dimension such as CMP number or shot number). The label of this field updates automatically to reflect the actual grouping key name from the input data. Changing this value navigates the display to the nearest matching gather on the location map.

Second

The secondary gather selector value, corresponding to the second grouping key of the input sorted headers (e.g., a second gather dimension for 3D data). Like the First parameter, its label updates automatically. Together, First and Second uniquely identify a gather when a two-dimensional gather index is used.

Velocity

The reference velocity (m/s) used to compute the linear moveout (LMO) reference curve overlaid on each gather display. The LMO curve shows where first-break arrivals should fall if the geometry is correct and the near-surface velocity equals this value. Default: 2000 m/s. Set this to the approximate near-surface refractor velocity of the survey area. A mismatch between the LMO curve and the auto-picked first breaks is a strong indicator of a geometry error in the corresponding gather.

Sliding params

A group of parameters controlling the interactive browsing window: how much of the trace time axis is displayed and how many gathers appear in the collection view at once.

Calculate maximum time automatically

When enabled (default: on), the vertical display window is scaled automatically to show all first-break arrivals and the LMO curve across the entire dataset. Disable this option to set the maximum display time manually using the Maximum time in window parameter below.

Maximum time in window

The maximum time (s) displayed in the gather view window. Active only when Calculate maximum time automatically is disabled. Default: 1 s. Valid range: 0.004 s to 20 s. Reduce this value to zoom in on the first-break zone and improve the visibility of geometry anomalies in shallow, fast surveys.

Window size X

The number of gathers to display side-by-side in the horizontal direction of the collection view. Default: 3. Minimum: 1. Increase this value to see more neighboring gathers simultaneously, which helps identify spatially localized geometry errors by comparison with surrounding gathers.

Window size Y

The number of gathers to display in the vertical direction of the collection view. Default: 3. Minimum: 1. Together with Window size X, these two parameters define the total number of gathers visible in the collection panel at any one time.

Hide empty values

When enabled, gathers that contain no valid traces (empty bins) are skipped in the collection view and are not shown as points on the location map. Default: off. Enable this option to declutter the display in surveys with sparse coverage.

Delay

A time delay (s) applied during the automatic slide-show playback triggered by the Start slide action. Default: 1 s. Increase this value to slow down the automatic browsing and allow more time to inspect each gather before advancing to the next.

Energy params

A group of parameters that control the automatic first-break picking algorithm. The picker uses a short-time energy characteristic function to detect the onset of the first arrival. These settings govern how that energy function is computed and how the resulting picks are cleaned up spatially.

Energy estimation window

The length of the sliding time window (s) used to compute the short-time energy characteristic function for first-break picking. Default: 0.05 s. Minimum: 0.001 s. A shorter window gives sharper onset detection but is more sensitive to noise; a longer window provides smoother picks but may smear the detected onset. Adjust based on the dominant frequency and noise level of your data.

Energy Median threshold

The relative amplitude threshold (fraction, 0 to 1) used in the spatial median filter applied to the energy-based first-break picks. Picks that deviate from the local spatial median by more than this fraction are flagged as outliers. Default: 0.1 (10%). Increase this value to accept more variation in the picks before flagging them; decrease it to apply stricter outlier rejection.

Energy Median filter aperture

The spatial radius (m) over which neighboring gathers are included when computing the local spatial median of energy-based first-break picks. Default: 300 m. Minimum: 1 m. A larger aperture produces smoother spatial statistics and is more robust to isolated noisy picks, but may reduce sensitivity to rapid near-surface variations. Set this to roughly the receiver line spacing or the expected scale of near-surface heterogeneity.

Time median parameters

A group of parameters for a secondary median filter applied in the time domain to the first-break pick times. This filter compares each pick to its spatial neighbors in pick-time space and can optionally exclude or replace outlying picks.

Time median threshold

The relative deviation threshold (fraction, 0 to 1) for the time-domain median filter on first-break pick times. Picks deviating from the local median by more than this fraction are considered anomalous. Default: 0 (filter disabled). Increase this value to activate the filter and control how strictly the pick times must follow the spatial trend.

Time median threshold

The absolute time deviation threshold (s) for the time-domain median filter. Picks whose first-break time differs from the local spatial median by more than this absolute amount are flagged as outliers. Default: 0.01 s. Valid range: 0 to 1 s. This provides an absolute safeguard complementary to the relative threshold above.

Time median filter aperture

The spatial radius (m) over which neighboring gathers are included when computing the local median of first-break pick times for the time-domain median filter. Default: 300 m. Minimum: 1 m. Use a larger aperture in areas with gradual near-surface velocity changes, and a smaller aperture where near-surface variability is high.

Time Median filter exclude

When enabled, picks identified as outliers by the time-domain median filter are completely excluded from the display and from the output table rather than being replaced by the median value. Default: off. Enable this option when you want to suppress anomalous picks entirely rather than smooth them.

Settings

Auto-connection

Controls whether the module automatically reconnects to its input data sources when the project is reopened. When enabled, the module restores its previous state and data connections without requiring manual re-linking.

SegyCacheParams

Controls how seismic trace data is cached in memory during interactive gather browsing. Tuning these settings can significantly improve the response speed when navigating through large datasets by keeping recently accessed gathers in memory.

Execute on { CPU, GPU }

Selects whether the first-break picking computation runs on the CPU or GPU. GPU execution can accelerate processing when working with a large number of gathers.

Distributed execution

Options for running the initial first-break picking pass across a cluster of processing nodes. When enabled, the gather-by-gather computation is distributed to reduce wall-clock time for large 3D datasets.

Bulk size

The minimum number of gathers assigned to each processing node or thread in a distributed or multi-threaded run. Larger values reduce inter-node communication overhead but may result in uneven load distribution at the end of the job.

Limit number of threads on nodes

Sets the maximum number of CPU threads each remote processing node may use. This prevents a single job from monopolizing all cores on a shared cluster node.

Job suffix

An optional text label appended to the job name in the distributed processing queue. Use this to distinguish multiple simultaneous runs of this module from one another in the cluster job manager.

Set custom affinity

When enabled, allows manual specification of the CPU core affinity for the processing threads, overriding the default operating system scheduling. Enable this only when precise CPU resource isolation is required.

Affinity

Specifies the CPU core mask or set to which processing threads are pinned when Set custom affinity is enabled.

Number of threads

The number of parallel CPU threads used during the initial first-break picking pass over all gathers. Increasing this value speeds up the pre-computation step on multi-core workstations. The default is typically determined automatically from the available hardware.

Skip

When enabled, the module execution is skipped in a processing flow without removing it from the chain. This is useful for temporarily bypassing the geometry check during batch reprocessing while preserving all parameter settings.

Output data

Output DataItem

The main seismic dataset container passed through to the next module. The trace data itself is not modified; this output carries the updated trace headers reflecting any geometry corrections applied interactively.

Output SEG-Y data handle

The SEG-Y file handle passed through for use by downstream modules.

Output trace headers

The output trace header table, which may contain corrected source or receiver coordinate values for traces whose geometry was identified and fixed during the interactive inspection session.

Output gather

The gather index passed through to downstream modules, reflecting the current state of the geometry after any corrections made during the interactive session.

Output stack line

The 2D stack line geometry passed through unchanged.

Output crooked line

The crooked-line geometry passed through unchanged.

Output bin grid

The bin grid geometry passed through unchanged.

Output sorted headers

The sorted gather index vector passed through to downstream modules.

Table changed points

A tabular record of all common-point gathers where geometry has been interactively modified during the session. Each row records the gather identifier and the corrected coordinate values. This table can be exported or used as input to a downstream geometry correction module to apply the identified fixes to the dataset.

Information

Graphics

Custom actions

Start slide

Starts an automatic slide-show that advances through all common-point gathers sequentially, pausing for the duration specified by the Delay parameter between each gather. Use this to rapidly scan the entire dataset for geometry anomalies without manually clicking through each gather.

Next

Advances the display to the next common-point gather in the sequence. Both the gather view and the location map highlight the newly selected gather.

Prev

Returns the display to the previous common-point gather in the sequence, allowing the operator to revisit a gather for closer inspection or to undo a geometry correction.