|
<< Click to Display Table of Contents >> Navigation: General > Convolutor |
Note: This module is deprecated. It is retained for compatibility with existing workflows but is no longer actively maintained. New projects should use the current modeling tools available in the Modeling group.
The Convolutor module performs convolutional seismic modeling. It takes a set of travel-time and reflectivity data (a trace travel-time vector) and convolves those reflection events with a synthetic wavelet to produce a simulated seismic gather. In essence, each reflection event in the input is replaced by a scaled and time-shifted copy of the chosen wavelet, yielding a synthetic seismogram.
When no travel-time input is connected and the Test wavelet only option is enabled, the module operates in wavelet-preview mode: it generates a small synthetic gather containing a single reflection placed at the centre of the record, allowing you to visualise the shape of the selected wavelet before running a full modeling job.
Typical use cases include building synthetic reference gathers for well-tie quality control, testing wavelet shapes before more complex finite-difference or ray-tracing modeling, and creating simplified seismic models for educational or feasibility purposes.
A travel-time vector item that describes the reflection geometry: for each output trace, it stores the two-way travel times and corresponding reflectivity coefficients at each reflection interface. This input is produced by ray-tracing or other modeling modules. If this item is not connected and Test wavelet only is enabled, the module will generate a simple single-spike gather instead of a full synthetic; if Test wavelet only is disabled and no travel-time data is connected, the module will return an error.
When enabled, the module ignores the travel-time input and instead generates a small test gather with a single reflection placed at the midpoint of the record. This mode is useful for previewing the wavelet shape and verifying that the wavelet parameters (frequency, length, type) produce the desired pulse before connecting real modeling data. Default: off.
The total number of time samples in each output trace. Together with the Sample rate parameter, this determines the total record length of the synthetic gather. For example, 800 samples at a 4 ms sample rate produces a 3.2-second record. Default: 800 samples. Minimum value: 1.
The time interval between consecutive samples in the output trace, expressed in seconds. This value is also referred to as the sample rate or sample interval. It controls the temporal resolution of the output synthetic gather and must be consistent with the sample interval of the travel-time input data. Default: 0.004 s (4 ms).
This parameter is defined in the module metadata but is not used in the current (deprecated) implementation of the Convolutor procedure. It was intended to control the amplitude level of random noise added to the synthetic gather, expressed as a percentage of the signal amplitude. If you require noise addition, use the Noiser module instead. Default: 10.
A parameter group that defines the shape of the synthetic wavelet used in convolution. It contains three sub-parameters:
Impulse type — selects the mathematical form of the wavelet to be convolved with the reflectivity series. The available wavelet types are: Ricker1 (Tong Fei's Ricker wavelet — the most common choice for synthetic seismograms), Ricker2 (Larner's variant of the Ricker wavelet), AKB (Akima wavelet), Berlage, Gaussian, GaussianDeriv (first derivative of a Gaussian), MinPhase, Klauder, Ormsby, Spike (single-sample impulse), and Zero (all zeros, useful for testing). For most well-tie and modeling workflows, Ricker1 is recommended.
Frequency — the dominant (peak) frequency of the synthetic wavelet, in Hz. Higher values produce shorter, sharper wavelets with better temporal resolution but reduced penetration. Lower values produce broader wavelets typical of deeper targets. Choose a value representative of the dominant frequency of the real seismic data you are modelling.
Wave length — the total duration of the wavelet window used during convolution, in seconds. This value determines how long a wavelet "ring" is applied at each reflection event. Set this value to cover at least two to three full cycles of the wavelet at the chosen dominant frequency. For a 30 Hz Ricker wavelet, a length of approximately 0.1 s (100 ms) is typically sufficient.