The Event autopicker module automatically refines an existing set of event picks stored in a trace header. It reads an initial pick time from a user-specified header field for each trace, then searches within a configurable time window around that pick to find the strongest local amplitude peak matching the expected polarity. The refined time is written to a new header field in the output gather, and a laterally smoothed version of the refined picks is written to a second new header field.
Use this module when you have a rough first-break or horizon pick already stored in a trace header and need to snap it to the nearest true amplitude peak on each trace. A typical workflow connects this module after a first-break picking or horizon picking step, allowing the raw picks to be automatically corrected before they are used for statics computation, NMO correction, or horizon interpretation. The three vista overlays (original picks, autopicked picks, and smoothed picks) are displayed simultaneously on the seismic gather view so you can evaluate the improvement quality before committing the result.
The module operates in three sequential stages. First it reads the initial pick time from the selected header for every trace and displays these as the "Events from data" overlay. Second it applies a local correlation-based peak-detection algorithm within the search window around each initial pick, selecting the peak whose correlation score is highest and whose polarity matches the selected event type; the result is displayed as the "Events from autopicker" overlay. Third it applies a median filter across the refined picks in the trace direction to suppress isolated outliers, producing the smoothed picks displayed as the "Events from autopicker smoothed" overlay. Both the autopicked and smoothed times are written as new custom headers in the output gather.
Input DataItem -
Connect the seismic gather that contains the initial event pick times stored as a trace header. This must be a gather in which each trace carries a numeric header value representing the time (in seconds) of the event you want to refine. The header field to use is selected using the Event header parameter. The gather may be a common-shot, CMP, or any other sorted domain — the autopicker works on a per-trace basis regardless of gather type.
Input gather -
The seismic gather connector. Both amplitude data and trace headers are read from this input. The amplitude data is used during the peak-search stage to identify the best pick position on each trace. The trace headers are read to obtain the initial event time from the field specified in the Event header parameter, and all existing headers are carried through to the output gather unchanged.
Event header -
Selects the trace header field that contains the initial event time for each trace. The dropdown is populated automatically when an input gather is connected, listing all standard SEG-Y headers (such as first-break picks written by upstream picking modules) as well as any custom headers attached to the data. The value in this header is treated as a pick time in seconds and serves as the starting point for the autopicking search window. If the header contains zero or an invalid value for some traces, those traces will receive the autopick at the center of the search window rather than failing.
Connect the input gather before configuring this parameter so that the dropdown is populated with the headers available in your dataset. Changing this parameter clears any previously calculated autopicks and forces a fresh computation on the next execution.
Event type { Positive, Negative } -
Specifies the polarity of the amplitude peak to snap to. Set to Positive (default) when your event of interest produces a positive amplitude excursion — for example, a first-break onset on land data or a strong positive reflection. Set to Negative when the event appears as a negative trough, which may occur depending on acquisition polarity conventions or the wavelet phase of the data.
Choosing the wrong polarity causes the autopicker to seek the wrong feature on each trace, resulting in systematically biased picks. Verify the polarity of your event in the seismic gather view before running. The default is Positive. Changing this parameter clears previously computed autopicks.
Search up time -
The amount of time, in seconds, to search above the initial pick time when locating the best amplitude peak. Together with Search down time, this defines the total search gate: the autopicker scans the window from (initial pick time - Search up time) to (initial pick time + Search down time) and selects the peak with the highest correlation score within that interval.
The default is 0.010 s (10 ms). Use a small value (5–20 ms) when your initial picks are already accurate to within one wavelet half-cycle; use a larger value (up to 50 ms or more) if the initial picks have significant timing errors. Setting this too large risks picking a strong nearby event that is not the target horizon. The minimum allowed value is 0.0 s. Changing this parameter clears previously computed autopicks.
Search down time -
The amount of time, in seconds, to search below the initial pick time when locating the best amplitude peak. This works symmetrically with Search up time to define the lower boundary of the search gate.
The default is 0.010 s (10 ms). You can set the upward and downward search ranges asymmetrically if your picks are known to be biased in one direction — for example, if first-break picks tend to be slightly early, you might set Search up time to 5 ms and Search down time to 20 ms. The minimum allowed value is 0.0 s. Changing this parameter clears previously computed autopicks.
SmoothWindow -
The half-width, in traces, of the median filter applied across the autopicked pick times to produce the smoothed pick curve. A window of N means the smoothed value at each trace is computed from the 2N+1 autopicked times centered on that trace. The median filter effectively removes isolated pick outliers (spikes caused by noise bursts or local amplitude anomalies) while preserving the general lateral trend of the picking horizon.
The default is 5 traces (giving an 11-trace smoothing window). Use smaller values (1–3 traces) when the near-surface model changes rapidly laterally and you want the smoothed picks to follow tight structural variations. Use larger values (10–50 traces) when you need a strongly regularized pick for statics application and short-wavelength variations are considered noise. The minimum allowed value is 0 (no smoothing). The smoothed pick is written to the Smoothed header name header field in the output gather.
Autopicker header name -
The name of the new custom trace header field that will be added to the output gather to store the autopicked (refined) event time for each trace. The default name is AUTOPICK_EVENT. You can change this to any string that is meaningful for your project — for example, FB_REFINED for a refined first-break pick. Downstream modules that consume pick times (such as Apply static shifts) should be configured to read from this header name.
Avoid using names that conflict with standard SEG-Y header fields. Changing this parameter name requires a full re-execution of the module because the output gather must be rewritten with the new header column. The autopicked times stored in this header are in seconds.
Smoothed header name -
The name of the new custom trace header field that will store the laterally smoothed version of the autopicked times. The default name is AUTOPICK_EVENT_SMOOTH. The smoothed picks are derived from the autopicked picks by applying a median filter across traces using the window width set in SmoothWindow. Use this header name in downstream statics or mute modules when you want to apply a regularized, outlier-free pick rather than the raw autopicked values.
Choose a name that clearly distinguishes the smoothed picks from the raw autopicks. Both headers are written to every trace in the output gather. Changing this name requires a full re-execution to rewrite the output gather.
Auto-connection - By default, TRUE(Checked).It will automatically connects to the next module. To avoid auto-connect, the user should uncheck this option.
Bad data values option { Fix, Notify, Continue } - This is applicable whenever there is a bad value or NaN (Not a Number) in the data. By default, Notify. While testing, it is good to opt as Notify option. Once we understand the root cause of it,
the user can either choose the option Fix or Continue. In this way, the job won't stop/fail during the production.
Notify - It will notify the issue if there are any bad values or NaN. This will halt the workflow execution.
Fix - It will fix the bad values and continue executing the workflow.
Continue - This option will continue the execution of the workflow however if there are any bad values or NaN, it won't fix it.
Number of threads - One less than total no of nodes/threads to execute a job in multi-thread mode. Limit number of threads on main machine.
Skip - By default, FALSE(Unchecked). This option helps to bypass the module from the workflow.
Output DataItem
The output gather is a copy of the input gather with two additional custom trace headers appended to every trace. The seismic amplitude data and all original trace headers are passed through unchanged. Only the two new header fields defined by Autopicker header name and Smoothed header name are added, containing the autopicked and smoothed pick times in seconds respectively.
Output gather -
The enriched seismic gather. This connector carries the same traces as the input but with the two new header fields populated with the autopicked and smoothed pick times. Connect this output to downstream modules that need to read the refined picks — for example, a statics application module that reads the AUTOPICK_EVENT_SMOOTH header to derive refraction statics corrections. Three vista point overlays are also produced and displayed on the seismic gather view for quality control: the original picks read from the input header, the raw autopicked picks, and the smoothed picks.
The Event autopicker has no custom action buttons. All computation is triggered automatically during normal workflow execution. To evaluate results interactively, view the seismic gather through the Points on seismic view vista group, which overlays the three pick curves (original, autopicked, and smoothed) directly on the amplitude display. Adjust the Search up time, Search down time, and SmoothWindow parameters and re-execute to iterate toward the best result.
YouTube video lesson, click here to open [VIDEO IN PROCESS...]
Yilmaz. O., 1987, Seismic data processing: Society of Exploration Geophysicist
* * * If you have any questions, please send an e-mail to: support@geomage.com * * *
