This section documents the processing steps available in the Filtering Steps category.
The processing steps currently available are:
|
|
Usage:
The 2D Damped Least Squares Filter step minimizes the weighted sum of (1) the deviation between the smooth velocity and the original one, and (2) the first derivative of the given signal. This minimization is equivalent to a second-order deferential equation that can be solved very efficiently.
Unlike conventional smoothing techniques (convolutional type) this method is independent of length of the smoothing operator, thus very efficient with long smoothing operators. This filter is very suitable for smoothing a 2D/3D velocity model before being used in migration or ray tracing. This step is multi-thread enabled.
Input Links:
1) Seismic data in any sort order (mandatory).
Output Links:
1) Seismic data in any sort order (mandatory).
References
Zhenyue L., A Velocity Smoothing Technique Based on Damped Least Squares, Colorado school of mines.
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Apply spatially If checked, the filter is applied in the horizontal direction.
Spatial smoothing operator Enter the degree of the smoothing operator in the horizontal direction.
Apply temporally (or depth) If checked, the filter is applied in the vertical direction.
Temporal smoothing operator Enter the degree of the smoothing operator in the vertical direction.
Define a window If checked, the smoothing will be applied only within the specified window
Window smoothing parameter Enter the smoothing operator value between the window and the rest of the record.
First trace first trace defining the window.
Last trace last trace defining the window.
First sample first sample defining the window.
Last sample last sample defining the window.
Usage:
The 2D Median Filter step is a single channel filter that may be used to remove spikes from your data or any other features, such as the wow in radar, which may be filtered by a median operator. You enter the side-length in samples and traces of the square window used for calculating the median value.
Input Links:
1) Seismic data in any sort order (mandatory).
Output Links:
1) Seismic data in any sort order (mandatory).
Reference:
-
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Window length in samples Enter the number of samples in the window over which the median will be calculated. The sample at the windows center will be replaced by the calculated median value.
Usage:
The 3D Diffusion Filter step is an anisotropic diffusion filter which performs edge-preserving smoothing in three dimensions. It is intended to be used to enhance post-stack 3D data.
The input data in the form of time traces does not require prior slicing or transformation.
Anisotropic diffusion is applied following structure in three dimensions using the Kuwahara filter. For this reason, estimation of the dip field is a required integral initial step. Parameters to control dip estimation are therefore required in addition to the diffusion parameters.
Due to parallel data handling, the output from 3D Diffusion Filter must be written directly to disk using Seismic File without further processes in that job flow.
Input Links:
1) 3D seismic data ordered by CMP Line and CMP Location (mandatory).
Output Links:
1) 3D seismic data ordered by CMP Line and CMP Location (mandatory).
Reference:
Fehmers, G.C, and Hocker, C.F.W., 2003, Fast structural interpretation with structure-oriented filters, Geophysics, vol. 68, no. 4, p. 1286-1293.
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Dip estimation paramters: define the parameters for dip estimation that is used for the edge preserving filter.
Maximum Dip (ms/trace) Enter the maximum dip to be considered in the filter.
Correlation window length (ms.) Enter the correlation window length for the dip scanning. An initial stage of dip estimation is cross-correlation of adjacent traces. This length should be sufficient to obtain stable dip estimates without being so long as to lose local structure changes.
Threshold coherency % Enter a normalised correlation values must exceed this % for the estimated dip at a point to be considered valid.
Use quadratic interpolation If checked, the dips are better estimated using a quadratic estimation formula.
Difussion parameters (Tukey): define parameters that allow distinguish noise from signal.
Sigma (%) Enter the noise threshold as a percentage of standard deviations
Lambda (%) Enter the percentage of new values to add on each filter iteration
Number of iterations Enter the number of filter iteration to apply to the input data.
Output dip field only If checked, only the estimated inline or cross-line dip field will be output. This is for diagnostic use: If the estimated dip field output is not smooth and contain spikes then the diffusion process will not be optimal. In this case the dip estimation parameters should be modified until a reasonable dip field is output.
In-line dip field (ms/trace) If checked, the dip field output will be the estimated in-line dip field.
Crossline dip field (ms/trace) If checked, the dip field output will be the estimated crossline dip field.
Usage:
The 3D Median Filter step is a filter with multiple uses which may be used for both single and multi-channel processing.
The input data in the form of time traces does not require prior slicing or transformation.
In Apply mode, 3D Median Filter replaces the input data samples with the median value of the surrounding values as specified by the filter size. In this mode the filter may be used to enhance post-stack 3D, common offset or pre-stack data by the application of short length median filters. For example, for post-stack 3D data the default filter 3-3-3 will replace the value of every sample with the median of its 27 surrounding samples.
In Remove mode, 3D Median Filter subtracts from the input data samples the median value of the surrounding values as specified by the filter size. In this mode the filter may be used to remove low frequency noise. For example the filter 1-1-41 will act as a trace independent low frequency removal filter, or 1-1-5 as a high frequency removal filter.
One and two dimension median filters in any orientation (using a length of one in the other dimension(s) will be quite fast. Filters which are long in all three dimensions are not recommended!
Input Links:
1) Seismic data in any sort order (mandatory).
Output Links:
1) Seismic data in any sort order (mandatory).
Reference:
-
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Apply median filter If selected, the data will be replaced by the median values as calculated by the specified filter. Use to enhance the data by removing random noise with a short filter in two or three dimensions.
Remove median filter If selected, the median values as calculated by the specified filter will be subtracted from the input data. Use to remove low frequency noise with a long filter in one or two dimensions
Filter size: define the size of the filter to be applied.
Width in traces Enter the number of traces within each record used to estimate median values.
Depth (records) in traces Enter the number of neighboring records used to estimate median values. Note: this assumes sufficient similarity of adjacent records such as use in post-stack 3d data.
Height in samples Enter the median filter length in the trace direction.
Usage:
The Adaptive Radon Demultiple step performs parabolic radon demultiple through a modeling of primaries, multiples, and noise in the parabolic radon domain followed by subtraction of multiples and noise in the time domain. You specify the transform type, the range of ray parameters in the output transform, a percentage of multiples plus noise to subtract, and the spatial and temporal taper lengths used to generate the transform.
Input Links:
1) Seismic data in any sort order (required).
Output Links:
1) Seismic data in any sort order (required).
Reference:
Hampson, D., 1991, Inverse velocity stacking for multiple elimination, Journal of the Canadian Society of Exploration Geophysics, 22, p. 44-55.
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Time window: Specify time parameters.
Limit application time If checked, the application time for the filter may be limited by using an input horizon or by using a constant time.
Start time referenced to horizon If selected, reference start time to a horizon for the filter.
Multiplier for time horizon Enter the multiplier for time horizon.
Use constant start time If seleceted, a constant time for limiting the application zone for the filter.
Start time (ms) Enter the start time in milliseconds for the application window of the filter.
Length of taper zone (ms) Enter the taper length for the transition zone for the windows. A Hanning taper is used.
Ray parameters: Specify ray parameters.
Set number of ray parameters If checked, the number of ray parameters to be calculated will be the number entered below. If not checked, the number of ray parameters will be equal to the number of traces in the input gather.
Number of ray parameters Enter the constant number of ray parameters to calculate for each gather.
Bandpass filter: Specify bandpass filter parameters.
Low cut (Hz) Enter the low cut frequency in Hertz for the bandpass filter.
Low pass (Hz) Enter the low pass frequency in Hertz for the bandpass filter.
High pass (Hz) Enter the high pass frequency in Hertz for the bandpass filter.
High cut (Hz) Enter the high cut frequency in Hertz for the bandpass filter.
Model moveout: Set the range of parabolas that will generate the primary and multiple models.
Use differential moveout If selected, specify the radon parameters in terms of differential moveout at far offset.
Minimum Enter the minimum differential moveout for calculation of the radon transform expressed in milliseconds on the far-offset trace. The primary model is generated with parabolas from the <minimum differential moveout> to the <minimum multiple moveout>.
Maximum Enter the maximum differential moveout for the calculation of the radon transform expressed in milliseconds on the far-offset trace. The multiple model is generated with parabolas from the <minimum multiple moveout> to the <maximum differential moveout>.
Minimum multiple Enter the minimum mulitple differential moveout for the radon filter expressed in milliseconds on the far-offset trace. All data between the minimum multiple differential moveout and the maximum differential moveout will be removed by the filter. The multiple model is generated with parabolas from the <minimum multiple moveout> to the <maximum differential moveout>.
Use velocity If selected. specify the radon filter parameterization using velocities.
Minimum moveout Enter the minimum (negative usually) velocity for the radon transform calculation.
Maximum moveout Enter the maximum velocity for the radon transform calculation.
Minimum multiple moveout Enter the minimum multiple moveout velocity for the radon filter. The filter will remove data between the minimum multiple moveout velocity and the maximum multiple moveout velocity.
Define maximum reference offset If checked, a constant maximum offset will be used for the calculation of the radon transform. The differential moveouts will be the delta times at this specified offset.
Maximum reference offset Enter a constant maximum offset to be used for the calculation of the radon transform. The differential moveouts will be the delta times at this specified offset.
Reference time (sec) Enter the reference time in seconds where the velocities are specified. This is used to convert the velocities to differential moveout for the radon algorithim.
Additional parameters: Specify additional parameters.
Percent add-back Enter the percentage of the original data to be blended with the output filtered data.
Percent pre-whitening Enter the percentage of pre-whitening to apply to the data for stabilizing the algorithm.
Minimum live traces Enter the minimum number of live traces per gather for applying the filter. Gathers with fewer traces will be passed through to the output with no filtering applied.
AGC window Set the parameters for an AGC window.
Apply and remove AGC If checked, apply an AGC before the filter and remove the same AGC after the filter.
AGC operator length (ms) Enter the AGC operator length in milliseconds.
Type of output If selected, the type of output desired.
Estimated primaries If selected, output the estimated primaries.
Multiple model Output the multiple data.
Example of Adaptive Radon Demultiple:
Usage:
The Apply F-K Filter step transforms T-X domain data to the F-K domain, applies a specified F-K reject filter, and returns the data to the T-X domain. F-K Filters are picked interactively as Surgical Mutes in SeisViewer using the Pick Traces tool located in the Picking menu.
Input Links:
1) Seismic data in any sort order (mandatory).
2) Surgical Mutes cards (mandatory).
Output Links:
1) Seismic data in same sort order as the input (mandatory).
Reference:
-
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Filter type: Define the type of F-K Filter.
Reject filter If selected, the polygon area is rejected.
Pass filter If selected, passes the polygon area.
Taper Type: Select the type of taper to use when applying the F-K mute function.
Hanning A Hanning taper is specified by the equation : x(n) = 0.5 - 0.5 * cos (2*pi*n/N).
Hamming A Hamming taper is specified by the equation : x(n) = 0.54 - 0.46 * cos (2*pi*n/N).
Blackman A Blackman taper is specified by the equation : x(n) = 0.42 - 0.5 * cos (2*pi*n/N)+ 0.08 * cos (4*pi*n/N)
No taper No taper will be applied to the mute. This may result in problems in later processing steps due to Gibbs effect.
Taper length Enter the mute taper length in samples. Longer taper lengths result in a smoother transition from the mute zone to the data zone.
AGC before filter If checked, an AGC is applied to each trace before filtering. The AGC gain function is removed after filtering.
Specify trace spacing If checked, specify manually the space between traces in record. Otherwise uses the trace spacing from headers. By default, the group interval is read from the seismic data and the trace-to-trace spacing is calculated from the group interval. If the Geometry Definition step has not been applied to the data, the seismic data will not contain information regarding the group interval and this option should be used.
Trace spacing (m or ft) Enter the trace interval in m or ft, depending on the project units.
Usage:
The Apply F-K Velocity Filter step transforms T-X domain data to the F-K domain, applies an F-K pass or reject filter, and returns the data to the T-X domain. A user-specified minimum and maximum velocity defines the pass or reject zone.
Input Links:
1) Seismic data in any sort order (mandatory).
Output Links:
1) Seismic data in any sort order (mandatory).
Reference:
-
Example Flowchart:
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Enter minimum velocity Enter the minimum velocity of the pass or reject zone.
Enter maximum velocity Enter the maximum velocity of the pass or reject zone.
Specify trace spacing If checked, manual set the trace-to-trace spacing in the record to be filtered. By default, the group interval is read from the seismic data and the trace-to-trace spacing is calculated from the group interval. If the Geometry Definition step has not been applied to the data, the seismic data will not contain information regarding the group interval and this option should be used.
Trace spacing (m or ft) Enter the trace interval in m or ft, depending on the project units.
Velocity type: Specify whether the applied F-K filter will be symmetrical or asymmetrical. See Figure below.
Double sided If selected, the specified minimum and maximum velocities will be used to create a double-sided, or symmetrical F-K filter.
Single sided (positive offsets) If selected, the specified minimum and maximum velocities will be used to pass or reject energy propagating in the direction of increasingly positive source-to-receiver offset.
Single sided (negative offsets) If selected, the specified minimum and maximum velocities will be used to pass or reject energy propagating in the direction of increasingly negative source-to-receiver offset.
Filter type: Specify whether the minimum and maximum velocity values specify a pass or a reject zone.
Reject filter If selected, the portion of F-K space defined by the minimum and maximum velocities will be rejected.
Pass filter If selected, the portion of F-K space defined by the minimum and maximum velocities will be passed.
Taper Type: Select the type of taper to use when applying the F-K mute function.
Hanning A Hanning taper is specified by the equation: x(n) = 0.5 - 0.5 * cos (2*pi*n/N).
Hamming A Hamming taper is specified by the equation: x(n) = 0.54 - 0.46 * cos (2*pi*n/N).
Blackman A Blackman taper is specified by the equation: x(n) = 0.42 - 0.5 * cos (2*pi*n/N)+ 0.08 * cos (4*pi*n/N)
No taper No taper will be applied to the mute. This may result in problems in later processing steps due to Gibbs effect.
AGC Window: Define parameter for AGC filter.
Apply and remove AGC If checked, and AGC filter is applied before the F-K Velocity filter and removed from the output seismic file. The AGC filter is used exclusively internally to improve the performance of the F-K filter.
AGC operator length Enter the AGC operator length in ms.
Taper length (ms) Enter the mute taper length in samples. Longer taper lengths result in a smoother transition from the mute zone to the data zone.
Usage:
The Butterworth Dip Filtering step is a single channel filter that may be used to pass or reject dipping data.
The data is transformed to the frequency domain, where the Buttterworth dip filter is applied.
Common usage includes suppression of dipping noise from pre-stack data.
Input Links:
1) Seismic data in any sort order (mandatory).
Output Links:
1) Seismic data in any sort order (mandatory).
Reference:
Hale, D. and Claerbout, J.F., 1983, Butterworth dip filters, Geophysics, v. 48, p. 1033-1038.
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Filter type: Define type of filter to be applied.
Pass If checked, the filter will pass dipping data over the specified range of dips.
Reject If checked, the filter will reject dipping data over the specified range of dips.
Dip specification: Specify parameters related with dip estimation
Milliseconds per trace If checked, the range of dips required must be input in terms of milliseconds per trace.
Velocity If checked, the range of dips required must be input in terms of apparent velocity.
Poles: Define number of poles in the filter.
Number of poles Enter the number of poles for the Butterworth dip filter. A larger number of poles for a sharper cut-off of the pass or reject zone.
Amplitude selection: Select type of amplitude.
Relative amplitude If checked, the data is equalized to relative amplitude before application of the filter.
True amplitude If checked, the filter is directly applied to the frequency domain data.
Frequency range: Specify the frequency range of the filter.
Maximum frequency Enter the maximum frequency to be used. The greatest frequency that is to be maintained and dip filtered.
Usage:
The Butterworth Filtering step allows you to apply recursively a Butterworth filter to your trace data in the time domain. You specify the low pass, high pass and high and low rolloff rates in decibels (dB) for the filter. You may choose to apply the filter as either a zero phase or minimum phase filter.
Input Links:
1) Seismic data in any sort order (mandatory).
2) Seismic data or auxiliary text file with filter (mandatory).
Output Links:
1) Seismic data in any sort order (mandatory).
Reference:
-
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Low-cut filtering: Specify low-cut filter parameters.
Apply low-cut filter If checked, a low-cut filter will be applied to your data.
Low-cut corner frequency (Hz) Enter the low half-power frequency in Hertz to be applied to your data. The amplitude of this frequency will be reduced by a factor of two relative to the input.
Low-cut rolloff rate (dB/oct) Enter the low rolloff filter slope in decibels per octave to be applied to your data. Higher numbers give steeper rolloff.
High-cut filtering: Specify high-cut filter parameters.
Apply high-cut filter If checked, a high-cut filter will be applied to your data.
High-cut corner frequency (Hz) Enter the high half-power frequency in Hertz to be applied to your data. The amplitude of this frequency will be reduced by a factor of two relative to the input.
High-cut rolloff rate (dB/oct) Enter the high rolloff filter slope in decibels per octave to be applied to your data. Higher numbers give steeper rolloff.
Phase selection: Select the phase type of the filter.
Zero Zero-phase filter selected.
Minimum Minimum-phase filter selected.
Usage:
The Coherence Enhancement Filter is used to accentuate the visual coherence of seismic events.
Input Links:
1) Seismic data in any sort order (mandatory).
Output Links:
1) Seismic data in any sort order (mandatory).
Reference:
-
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Coherence Exponent Enter the coherency filter exponent. Values of 1 to 1.5 are usually used. A value of 1 does nothing. A value of 1.5 significantly increases coherency. The exponent is applied to the data in the F-K domain.
Usage:
The Convolution step is used to convolve traces in a seismic data file with a filter function specified in an auxiliary data set define as seismic file or auxiliary text file.
Input Links:
1) Seismic data in any sort order (mandatory).
2) Auxiliary file in any sort order (mandatory).
Output Links:
1) Seismic data in any sort order (mandatory).
Reference:
-
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Filter location: Specify the location of the filter to be convolved.
Seismic file If selected, the filter is defined in an auxiliary seismic file.
Text file If selected, the filter is defined in an auxiliary text file.
Usage:
The Cross Correlation step computes the cross correlation between an input seismic data and an auxiliary seismic data.
Input Links:
1) Seismic data in any sort order (mandatory).
2) Auxiliary Seismic data in any sort order (mandatory).
Output Links:
1) Seismic data in any sort order (mandatory).
Reference:
-
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Input data: Specify parameters for input seismic file.
Window start time Enter the window start time for input seismic file in milliseconds.
Window end time Enter the window end time for input seismic file in milliseconds.
Auxiliary data: Specify parameters for input seismic file.
Window start time Enter the window start time for auxiliary seismic file in milliseconds.
Window end time Enter the window start time for auxiliary seismic file in milliseconds.
One input trace per record If selected, the auxiliary data is considered to have one input trace per record.
One input trace per dataset If selected, the auxiliary data is considered to have one input trace per record.
Zero lag time Enter the zero lag time for the cross correlation.
Output length (ms) Enter the output length of the cross correlation in milliseconds.
Usage:
The Derivative step computes the derivative of the data samples in each input seismic trace.
Input Links:
3) Seismic data in any sort order (mandatory).
Output Links:
2) Seismic data in any sort order (mandatory).
Reference:
-
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
There are no defined parameters for this step.
Usage:
The Diffusion Filter is a noise reduction filter based on the diffusion equation. The filter performs both smoothing and edge preservation.
The most common usage is to enhance post-stack 3D data by applying Diffusion Filter to time slices.
Use Time Slice (Processing Category Display) in a prior job flow to convert such data to time slices for input to Diffusion Filter.
After Diffusion Filter, use Time Slice to Trace (Processing Category Display) in a further job flow to revert the filtered data to normal order.
An alternative to Diffusion Filter for enhancing post-stack 3D data is use of 3D Diffusion Filter.
Input Links:
1) Seismic data in any sort order (mandatory).
Output Links:
1) Seismic data in any sort order (mandatory).
Reference:
Fehmers, G.C, and Hocker, C.F.W., 2003, Fast structural interpretation with structure-oriented filters, Geophysics, vol. 68, no. 4, p. 1286-1293.
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Difussion method: Define the diffusion method.
Structure gradient diffusion If selected, use structure-oriented filtering (Fehmers & Hocker)
Tukey diffusion If selected, use an anisotropic diffusion using the Kuwahara filter
Structure gradient diffusion: specify the structure gradient diffusion.
Gradient scale Enter the number of standard deviation of 2D Gaussian filter used for smoothing the input data prior to gradient calculation.
Structure scale Enter the number of standard deviation of 2D Gaussian filter used for smoothing the gradient tensors prior to eigendecomposition.
Number of iterations Enter the number of filter iteration to apply to the input data.
Use Edge-preserving filters If checked, a continuity (coherence) factor is calculated that inhibits diffusion across edges.
Sigma (%) Noise threshold as a percentage of standard deviations
Lambda (%) Percentage of new values to add on each filter iteration
Number of iterations Number of filter iteration to apply to the input data.
Usage:
The F-K Spectrum step allows you to create F-K spectrum image plots of your seismic data for designing F-K filters. Alternatively, the F-K spectrum can be directly created in SeisViewer from a selected seismic file.
Input Links:
1) Seismic data in any sort order (mandatory).
Output Links:
1) Seismic data containing F-K spectrum (mandatory).
Reference:
-
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Input amplitude treatment: The amplitude to use as input to the F-K transform, either relative or true amplitude
True amplitude If selected, use the true seismic amplitude as in the input seismic traces for the analysis. True amplitude traces are scaled by one common factor per record.
True amplitude If selected, use relative seisimc amplitude scaled traces in the analysis. Relative amplitude traces are scaled independently of one another.
Output amplitude scale: Select the output amplitude scale, either absolute amplitude or dB down from the maximum.
Amplitude If selected, the output amplitude scale is set absolute amplitude.
Decibel If selected, the output amplitude scale is set in dB down from the maximum.
Specify trace spacing If checked, specify manually the space between traces in record. Otherwise uses the trace spacing from headers. By default, the group interval is read from the seismic data and the trace-to-trace spacing is calculated from the group interval. If the Geometry Definition step has not been applied to the data, the seismic data will not contain information regarding the group interval and this option should be used.
Trace spacing (m or ft) Enter the trace interval in m or ft, depending on the project units.
dB scale control: Set zero dB amplitude for comparison of two dB scale iamges.
Set 0dB value If checked, the zero dB amplitude value may be entered. Using this option allows you to compare two dB scale images.
Zero dB value Enter the Zero dB value and its exponent.
Usage:
The F-X Deconvolution step is a 2D multi-channel noise filter designed to attenuate random noise. This time-variant, adaptive filter removes the non-predictable part of the data using the assumption that the signal portion of the data is predictable and the noise portion of the data is inherently random, and therefore non-predictable. You specify the length of the filter window, in samples, the width of the filter window, in traces, and the filter adaptation percent.
Input Links:
1) Seismic data in any sort order (mandatory).
Output Links:
1) Seismic data in any sort order (mandatory).
Reference:
Hornbostel, S., 1991, Spatial prediction filtering in the t-x and f-x domain, Geophysics, v. 56, no 12, p. 2019-2026.
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Design window length: Specify the length of the window of the filter.
Whole trace If checked, the length of the filter is equal to the length of the input seismic trace.
Specify length If checked, the length of the filter is defined manually.
Length (ms) Enter the length of the filter in ms.
Taper length (ms) Enter the taper length of the filter in ms.
Design window width: Specify the width of the filter.
Entire record If checked, uses the width filter is equal to the number of traces of the input record.
Specify width If checked, uses the width filter is manually defined.
Width (traces) Enter the width of the filter in number of traces.
Prediction filter: Specify the size of the prediction filter.
Width (traces) Enter the width of the prediction filter in number of traces.
Processing Frequency Range: Specify the frequency band where the filter is going to be applied.
Minimum frequency (Hz.) Enter the minimum frequency value in Hz of the frequency band.
Maximum frequency (Hz.) Enter the maximum frequency value in Hz of the frequency band.
Usage:
The F-X Median Filter step is a single channel filter that may be used to remove spikes or other noise from your data without damaging dipping data. The operator is applied in the temporal frequency domain using fully overlapping cosign-tapered window panes of specified height. A horizontal 1D median filter is applied to the frequency amplitudes whilst leaving the phase unchanged. Either the entire dataset or a specified tapered time window may be filtered
The input data in the form of time traces and does not require prior slicing or transformation.
Input Links:
1) Seismic data in any sort order (mandatory).
Output Links:
1) Seismic data in any sort order (mandatory).
Reference:
-
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Paramters: Specify the median filter parameters.
Filter width in traces Enter the width in number of traces of the spacial median filter to be applied to each temporal frequency component.
Overlapping panes height (ms.) Enter the height in milliseconds of each of the fully-overlapping cosine-tapered panes from which frequency components are derived. In general, shorter panes will exhibit more filtering, but these should be sufficiently long to allow realistic longer frequency estimation.
Data selection: Specify the size of the data input for the median filter.
All data If checked, the entire data length will be filtered.
Specified extent If checked, the specified data aperture will be filtered.
Start time (ms.) Enter the required time at which filtering is to begin.
End time (ms.) Enter the required time at which filtering is to end.
Taper length (ms.) Enter the length of the start and end tapers used to taper in the filtered data to the input data.
Usage:
The F-X Noise Attenuation step is a random noise attenuation technique in the frequency-space domain. This filtering method predicts the signal with spatial changes in dir or amplitude using the F-X domain.
Input Links:
1) Seismic data in any sort order (mandatory).
Output Links:
1) Seismic data in any sort order (mandatory).
Reference:
-
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Design window length (ms) Enter the filter length in ms.
Design window width (traces) Enter the width of the filter in number of traces.
Filter order Enter the filter order.
Filter adaptation (%) enter the filter adaptation in percentage.
Usage:
The F-X-Y Median Filter step is a multi-channel filter that may be used to remove spikes or other noise from your data without damaging dipping data. The operator is applied in the temporal frequency domain using fully overlapping cosign-tapered window panes of specified height. A horizontal 2D median filter is applied to the frequency amplitudes whilst leaving the phase unchanged. Either the entire dataset or a specified tapered time window may be filtered.
The input data in the form of time traces and does not require prior slicing or transformation.
Note: This filter will be most often used with 3D post-stack data sorted by CMP line and CMP location, but it may be used with other data provided adjacent records are sufficiently similar for median filtering across records to make sense.
Due to parallel data handling, the output from F-X-Y Median Filter must be written directly to disk using Seismic File without further processes in that job flow.
Input Links:
1) Seismic data in any sort order (mandatory).
Output Links:
1) Seismic data in any sort order (mandatory).
Reference:
-
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Parameters: Specify the filter parameters.
Filter width in traces Enter the width of the spacial median filter to be applied to each temporal frequency component.
Overlapping panes height (ms.) Enter the height in milliseconds of each of the fully-overlapping cosine-tapered panes from which frequency components are derived. In general, shorter panes will exhibit more filtering, but these should be sufficiently long to allow realistic longer frequency estimation.
Data selection: Specify data to be used to define the filter.
All data If checked the entire data length will be filtered.
Specified extent If checked the specified data aperture will be filtered.
Start time (ms.) Enter the required time at which filtering is to begin.
End time (ms.) Enter the required time at which filtering is to end.
Taper length (ms.) Enter the length of the start and end tapers used to taper in the filtered data to the input data.
Usage:
The Footprint Filter step is a KK filtering technique for application to 3D post-stack data that is focused on attenuating acquisition footprint noise.
Two techniques are available:
1) Application of a global high cut KK filter. Seismic signal generally displays most of its energy in the KK domain close to zero, whereas the jitter sometimes seen due to acquisition will have high KK amplitudes. This filter is then simple, effective and fast.
2) A local adaptive noise removal algorithm for when the acquisition footprint exhibits in the KK domain as spikes. The seismic signal energy is protected with an exclusion zone around the origin. This technique is significantly more time-consuming and should be used only when the first technique does not address the type of acquisition noise seen.
Footprint Filter is applied to post-stack 3D data which has been converted to time slices.
Use Time Slice (Processing Category Display) in a prior job flow to convert such data to time slices for input to Footprint Filter.
After Footprint Filter, use Time Slice to Trace (Processing Category Display) in a further job flow to revert the filtered data to normal order.
Input Links:
1) Seismic data sorted by Slice Number (record key) and CMP Line (mandatory).
Output Links:
1) Seismic data sorted by Slice Number (record key) and CMP Line (mandatory).
Reference:
-
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Apply high-cut KK filtering If checked, a high-cut KK cosine filter as specified will be applied.
High-cut KK filtering: Specify the high-cut KK filter.
High pass frequency (% Nyq) Enter the KK taper-off beginning frequency.
High cut frequency (% Nyq) Enter the KK taper-off ending frequency.
Apply adaptive noise removal If checked, an adaptive noise removal filter will be applied.
Adaptive noise removal: Specify the adaptive noise removal parameters.
Noise threshold Enter the number of standard deviations of a sample amplitude above the standard deviation of the sample amplitudes within a local area or patch for that sample to be regarded as noise.
Patch width Enter the KK plane is divided into overlapping patches to be analyzed for noise spikes. This parameter is the side length of each square patch.
Number of iterations Enter the number of iterations of the adaptive noise removal algorithm.
Removal method: Select the removal method for the KK filter.
Replace noise samples If selected, the amplitude of a noise sample will be replaced by the average value within the patch.
Remove noise samples If selected, the amplitude of a noise sample will be set to zero.
Filtering signal exclusion zone: Specify a signal exclusion zone based on the percentage of the Nyquist.
Exclusion zone size (% Nyq) Enter the size of the area around the KK origin to be protected from the adaptive filter so as to preserve the signal.
Mask dead areas If checked, then input data dead zones will be masked after application of the KK filtering.
Usage:
The Horizontal Median step allows you to apply a horizontal median filter across the data traces of a gather or stack. This process is very useful in processing VSP data for separation of up-going and down-going events.
Input Links:
1) Seismic data in any sort order (mandatory).
Output Links:
1) Seismic data in any sort order (mandatory).
References:
Hardage, B. A., 1983, Vertical Seismic Profiling, Geophysical Press.
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Window length in traces Enter the number of traces in the spatial window.
Use quantile for median application If checked, the median value is computed over the selected quantile.
Quantile value Enter the quantile value to calculate the median.
Apply for VSP upgoing traces only If checked, the horizontal median filter is only applied to seismic traces marked as VSP upgoing traces.
Usage:
The Integration step integrates data samples in each input seismic data trace.
Input Links:
1) Seismic data in any sort order (mandatory).
Output Links:
1) Seismic data in same sort order as the input (mandatory).
Reference:
-
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
There are no parameters for this step.
Usage:
The Notch Filter step applies a frequency domain notch filter to the input data. The filter is specified by describing the frequency notch and the width of the notch. Options exist to (1) apply the filter as a function of header word flags, (2) parameterize the notch as a function of header word flags.
Input Links:
1) Seismic data in any sort order (mandatory).
Output Links:
1) Seismic data in any sort order (mandatory).
Reference:
-
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Application window: Specify the window of application for the notch filter.
Apply filter based on a trace header flag If checked, the application of the filter is to be controlled by the value of a trace header flag.
Trace header field Enter the value of a trace header flag, select the trace header from the drop down menu.
Trace header value Enter the value of a trace header flag, set the value of the trace header. When the value of the trace header equals the specified value, the filter will be applied. Otherwise, the filter will not be applied.
Notch frequency is in a trace header field If checked, the value of the notch frequency is to be controlled by the value of a trace header field.
Trace header field Enter the value of a trace header field, select the trace header from the drop down menu.
Notch frequency (Hz) Enter the value of the notch frequency in Hertz.
Notch width (Hz) Enter the width of the notch frequency in Hertz.
Usage:
The Quadrature Filter step computes the Hilbert transform of the input seismic data. Quadrature filters are usually applied to enhance the presence of discontinuities and reveal the structure of the seismic image.
Input Links:
1) Seismic data in any sort order (mandatory).
Output Links:
1) Seismic data in any sort order (mandatory).
Reference:
-
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Processing start time (ms) Enter the processing start time in ms.
Processing end time (ms) Enter the processing start time in ms.
Usage:
The Radial Transform Filter step is a 2D pre-stack single channel filter for attenuation of noise originating from a point source with a specific apparent velocity by means of the radial transform.
For example, shot generated noise may be attenuated by application of Radial Transform Filter to common shot gathers.
Each input record must be ordered by offset.
The input 2D record is first converted to the radial domain such that each radial trace represents a different apparent velocity within the range of interest, radiating from a start time (usually zero) at zero offset.
Using the output data type Radial traces, these traces may be output and displayed:
Coherent noise of particular apparent velocity will appear as low frequency in this domain, and can be attenuated by use of a low frequency filter or a median reject filter.
Uncontaminated data remains untouched by the radial transform. The region of the input data records which will be filtered may be displayed using the output data type Filter region model:
Input Links:
1) 2D Seismic data offset-sorted records (mandatory).
Output Links:
1) 2D Seismic data offset-sorted records (mandatory).
Reference:
Henley, D., 2008, A Convenient Truth: Radial Trace Filtering Simple and Effective, CSEG Recorder, vol. 34, no 1.
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Transform start time (ms) Enter the time of the origin for the radial transform. This will often be zero, but sometimes a positive value can give better noise attenuation.
Output data type: Specify type of output of the Radon Transform.
Filtered data If selected, the output data will be Radial Transform filtered data.
Radial traces If selected, the output will be the radial transformed traces, filtered or unfiltered depending on the parameter settings (See Usage above).
Filter region model If selected, the output will allow display of the exact region of the input data which will be filtered by Radial Transform Filter in its Filtered data mode (See Usage above).
Filter the radial traces If checked, the radial traces will be filtered.
Radial trace filtering: Specify type of rejection filter.
Apply a low cut frequency If selected, a low cut filter will be applied.
Low cut filter: Specify low cut parameters
Low cut frequency Specify the low cut frequency of the Butterworth filter.
Reject dB/octave Specify the rate of fall-off of the filter.
Apply a median reject filter If selected, a median reject filter will be applied.
Median reject filter: Specify median reject filter parameters. Low frequencies are attenuated by subtracting the median filtered data from the data.
Lowest freq to protect Specify the lowest frequency that will not be attenuated by the median reject filter. The median filter length used will be the wavelength of this frequency in samples.
Mute the radial traces If checked, the radial traces will be attenuated.
Radial trace mute: Specify the radial trace mute parameters.
dB reduction All data within the radial traces will be attenuated by the specified amount.
% tapering Specify the percentage of the radial trace zone(s) which will be tapered.
Interpolation type: Deriving the radial traces from the input data requires interpolation of the data. Nearest offset is the fastest, while linear and quadratic interpolation may give a smoother result and better radial domain low frequency attenuation.
Nearest offset If selected, radial traces will be constructed by picking nearest offset samples.
Linear If selected, linear interpolation will be used to construct radial traces.
Quadratic If selected, quadratic interpolation will be used to construct radial traces.
The above parameters may be adjusted by trial and error to obtain the best attenuation, or by inspection of the radial traces.
Usage:
The Radon Demultiple step performs parabolic radon demultiple through a modeling of multiples in the parabolic radon domain followed by subtraction of those multiples in the time domain. You specify the transform type, the range of ray parameters in the output transform, and the spatial and temporal taper lengths used to generate the transform.
Input Links:
1) Seismic data in any sort order (mandatory).
Output Links:
1) Seismic data in any sort order (mandatory).
Reference:
Hampson, D., 1991, Inverse velocity stacking for multiple elimination, Journal of the Canadian Society of Exploration Geophysics, 22, p. 44-55.
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Time window:
Limit application time
Use constant start time
Start time (ms)
Start time referenced to horizon
Multiplier for time horizon
Length of taper zone (ms)
Model moveout: Set the range of parabolas that will generate the primary and multiple models.
Minimum differential moveout Enter the minimum differential moveout, expressed in milliseconds on the far-offset trace. The primary model is generated with parabolas from the <minimum differential moveout> to the <minimum multiple moveout>.
Maximum differential moveout Enter the maximum differential moveout, expressed in milliseconds on the far-offset trace. The multiple model is generated with parabolas from the <minimum multiple moveout> to the <maximum differential moveout>.
Minimum multiple moveout Enter the minimum multiple moveout, expressed in milliseconds on the far-offset trace. The multiple model is generated with parabolas from the <minimum multiple moveout> to the <maximum differential moveout>.
Ray parameters: Specify ray parameters
Set number of ray parameters If checked, the number of ray parameters is determined manually. By default, the number of ray parameters is determined internally.
Number of ray parameters set the number of ray parameter.
Offset window: Speficy paramters for window definition.
Define maximum reference offset If checked, limit the input seismic data by a maximum reference offset
Maximum reference offset Enter the reference maximum offset. Maximum offset of the input record that is going to be used.
AGC Window: Define parameter for AGC filter.
Apply and remove AGC If checked, and AGC filter is applied before the F-K Velocity filter and removed from the output seismic file. The AGC filter is used exclusively internally to improve the performance of the F-K filter.
AGC operator length Enter the AGC operator length in ms.
Bandpass filter: Specify bandpass filter parameters, as defined in a trapezoid, to be applied during the processing.
Low cut (Hz) enter low cut frequency in Hz.
Low pass (Hz) enter low pass frequency in Hz.
High pass (Hz) enter high pass frequency in Hz.
High cut (Hz) enter high cut frequency in Hz.
Aditional parameters: Specify additional parameters to stabilize the Randon Demultiple.
Percent pre-whitening Enter the amount of pre-whitening used to stabilize the least-squares inversion in the presence of noise.
Minimum live traces Enter the minimum number of live traces that must be present in a gather in order to transform that gather.
Type of output: Specify type of output form Radon Demultiple.
Estimated primaries If selected, output estimated primary reflections.
Multiple model If selected, output modelled multiple reflection. This multiple model can then be subtracted to the input seismic data.
Usage:
The Ricker Filter step applies a ricker filter to the input seismic file. The Ricker filter is completely determined by the center frequency of the Ricker wavelet.
Input Links:
1) Seismic data in any sort order (mandatory).
Output Links:
1) Seismic data in any sort order (mandatory).
Reference:
-
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Center frequency (Hz) Specify the center frequency of the Ricker wavelet.
Usage:
The Swell Statics Correction step is a moving average filter designed for marine data used to attenuate statics originated by marine swell.
Input Links:
1) Seismic data in any sort order (mandatory).
Output Links:
1) Seismic data in any sort order (mandatory).
Reference:
-
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Traces in design window Enter the number of traces to be used in the window filter.
Maximum allowable shift (samples) Enter the maximum allowable shift, in number of seismic samples, that a sample may suffer while being filtered.
Usage:
The Tau-P Inverse Transform step transforms 2D Tau-P domain pre-stack data back to the time-offset domain after conversion to this domain after the use of Tau-P Transform.
Input Links:
1) 2D Seismic pre-stack Tau-P domain data records ordered by Ray Parameter (mandatory).
2) Auxiliary seismic data file: the original offset-time shot records for restoration of headers to the inverted data.
Output Links:
1) 2D Seismic pre-stack data records ordered by offset (mandatory).
Reference:
Kostov, C., 1990, Toeplitz Structure in Slant-Stack Inversion: 60th Annual Internat..Mtg., Soc. Expl. Geophys., Expanded Abstracrs, 1618-1621.
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Type of Tau-P transform: Specify the function of offset used for the Tau-P transform.
Parabolic If selected, a parabolic inverse transform will be performed. This must be the same as was used for the forward transform.
Linear If selected, a linear inverse transform will be performed. This must be the same as was used for the forward transform.
Output data offsets: Specify whether to use the original t-x data to obtain the offsets to be inverted, or to specify regular offsets.
Original t-x data offsets If selected, an auxiliary input file will be required. This should be the original t-x data before forward Tau-P transform. The offsets returned will match those on this file.
Regular offsets If selected, the inverse transformed records will contain regular offsets.
Bandpass filter: A bandpass filter will be applied to the data, before inverse Tau-P transformation. It is sensible if the same frequency range is inverted as specified for the forward transform, but not essential. If aliasing noise is a problem on inversion, then the inversion high pass or high cut frequencies may be decreased.
Low cut (Hz) Enter the bandpass low cut frequency in Hz.
Low pass (Hz) Enter the bandpass low pass frequency in Hz.
High pass (Hz) Enter the bandpass high pass frequency in Hz.
High cut (Hz) Enter the bandpass high cut frequency in Hz.
Percent pre-whitening Enter the percentage of pre-whitening. A Toeplitz matrix inversion is used for the Tau-P inversion of each frequency. A small amount of white noise added may be needed to stabilize the inversion. A fraction of one percent is often sufficient, and decreases the smearing sometimes seen when larger values are used.
Usage:
The Tau-P Transform step transforms 2D pre-stack data to the Tau-P domain for further processing. The Tau-P domain data may be inverted to the time-offset domain by the use of Tau-P Inverse Transform.
The input traces of each record are time-shifted according to their offsets, and then a low-cut spatial frequency filter is applied. This will attenuate much of the noise energy at that particular apparent velocity. The time-shifted data is then restored.
Input Links:
1) 2D Seismic pre-stack data records ordered by offset (mandatory).
Output Links:
1) 2D Seismic pre-stack Tau-P domain data records ordered by Ray Parameter (mandatory).
Reference:
Kostov, C., 1990, Toeplitz Structure in Slant-Stack Inversion: 60th Annual Internat..Mtg., Soc. Expl. Geophys., Expanded Abstracrs, 1618-1621.
Example Flowchart:
Step Parameter Dialog:
Parabolic transform parameters:
Linear transform parameters:
Parameter Description:
Type of Tau-P transform: Specify the function of offset used for the Tau-P transform.
Parabolic If checked, a signed offset squared function will be used. This transform type yields the best inversion back to the original domain
Linear If checked, the function used will be offset itself.
P values specification: Specify which method will be used to specify the range of P values to be calculated.
Use Velocity If checked, the range of P values to be calculated will be determined by the minimum and maximum velocities specified. In the case of the Parabolic transform, these velocities will be moveout velocities. In the case of the Linear transform, these velocities will be apparent velocities.
Specify a differential moveout range If checked, the minimum P value will be calculated from the maximum time shift at the specified max offset. The maximum P value will be calculated from the minimum time shift at the specified max offset.
Split-spread (-ve offsets)? If checked, the P-values calculated will be a negative P-value range and a positive P-value range.
Minimum live traces Enter the minimum number of live traces read for a record to be transformed to the Tau-P domain.
Velocity P values range specification: Specify the range of P values to be calculated by means of velocity.
Minimum moveout velocity (Parabolic transform) Enter the minimum moveout velocity. The maximum P value will be calculated from this and the specified reference time.
Maximum moveout velocity (Parabolic transform) Enter the maximum moveout velocity. The minimum P value will be calculated from this and the specified reference time.
Minimum apparent velocity (Linear transform) Enter the minimum apparent velocity. The maximum P value to calculate will be the inverse of this velocity.
Maximum apparent velocity (Linear transform) Enter the maximum apparent velocity. The minimum P value to calculate will be the inverse of this velocity.
Reference time (sec) (Parabolic transform only). Enter the time in seconds at which the moveout velocity is measured.
Time P values range specification: Specify the range of P values to be calculated by means of differential moveout.
Min differential moveout (ms.) (Parabolic transform) Enter the minimum differential moveout at the specified reference time. The minimum P value to calculate will be calculated from this value.
Max differential moveout (ms.) (Parabolic transform) Enter the maximum differential moveout at the specified reference time. The maximum P value to calculate will be calculated from this value.
Min slope (ms./1000) (Linear transform) Enter the minimum slope in milliseconds per thousand metres/feet. The minimum P value to calculate will be calculated from this value.
Max slope (ms./1000) (Linear transform) Enter the maximum slope in milliseconds per thousand metres/feet. The maximum P value to calculate will be calculated from this value.
Reference max offset (Parabolic transform only) Enter the offset for the minimum and maximum differential moveouts specified.
Ray parameters: Specity ray parameters.
Number of ray parameters Enter the number of P values to be calculated. If split-spread, half of these will be used for negative P values and half for positive. The number required depends on the total range of ray parameters. These P-values will be regularly spaced. If the increment multiplied by the offset is greater than half a wavelength of a frequency, then aliasing noise on inverse transform may result. The inversion execution time depends on this parameter, and testing with a few shot records is recommended to determine the smallest value possible without undue aliasing noise.
Bandpass filter: Both the forward and inverse Tau-P transforms perform calculation in the frequency domain. A bandpass filter is applied to the data, and the frequency range specified here should be based on the signal spectrum and minimization of aliasing noise.
Low cut (Hz) Enter the bandpass low cut frequency in Hz.
Low pass (Hz) Enter the bandpass low pass frequency in Hz.
High pass (Hz) Enter the bandpass high pass frequency in Hz.
High cut (Hz) Enter the bandpass high cut frequency in Hz.
Usage:
The Time Variant Bandpass step allows you to apply up to five (5) different time-variant bandpass filters to your trace data. You specify the low cut, low pass, high pass and high cut, filter points for each filter and the starting time for application of the filter. You also specify the filter taper length, which allows you to control the smoothness of the transition between adjacent filters.
Input Links:
1) Seismic data in any sort order (mandatory).
Output Links:
1) Seismic data in any sort order (mandatory).
Reference:
-
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Number of filters to use Enter the number of filters to apply.
Filter taper length Enter the length of the filter taper in samples. The longer the filters taper length, the smoother the transition between adjacent filters.
Low cut Enter the low cut frequency of the bandpass filter in Hertz.
Low pass Enter the low pass frequency of the bandpass filter in Hertz.
High pass Enter the high pass frequency of the bandpass filter in Hertz.
High cut Enter the high cut frequency of the bandpass filter in Hertz.
Start time (ms) Enter the start time in milliseconds to start the application of each filter.
Usage:
The Time Variant Butterworth step allows you to apply up to five (5) different time-variant Butterworth filters to your trace data. You specify the low pass, high pass and low and high rolloff rates in decibels (dB) for each filter, as well as the starting time for application of the filter. You also specify the filter taper length, which controls the smoothness of the transition between adjacent filters.
Input Links:
1) Seismic data in any sort order (mandatory).
Output Links:
1) Seismic data in any sort order (mandatory).
Reference:
-
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Number of filters to use Enter the number of filters to apply.
Filter taper length Enter the length of the filter taper in samples. The longer the filters taper length, the smoother the transition between adjacent filters.
Low frequency Enter the low pass frequency of the Butterworth filter in Hertz.
Low rolloff rate Enter the low pass rolloff rate in dB/Octave. Higher numbers give a steeper filter rolloff.
High frequency Enter the high pass frequency of the Butterworth filter in Hertz.
High rolloff rate Enter the high pass rolloff rate in dB/Octave. Higher numbers give a steeper filter rolloff.
Start Time (ms) Enter the start time in milliseconds to the start application of each filter.
Usage:
The Velocity-Guided Butterworth Filter step is a 2D pre-stack single channel filter for Butterworth filte with a particular apparent velocity in the offset-time domain.
The input traces of each record are time-shifted according to their offsets, and then a low-cut spatial frequency filter is applied. This will attenuate much of the noise energy at that particular apparent velocity. The time-shifted data is then restored.
Input Links:
1) 2D Seismic pre-stack data records ordered by offset (mandatory).
Output Links:
1) 2D Seismic pre-stack data records ordered by offset (mandatory).
Reference:
-
Example Flowchart:
Step Parameter Dialog:
Parameter Description:
Velocity cone: Specify velocity parameters.
Linear moveout velocity filter Enter linear moveout velocity for the Butterworth filter in m/s or ft/s.
Start time at zero-offset (ms) Enter the start time at zero-offset in milliseconds
Filter type: Specify filter type.
Apply inside cone If selected, apply the Butterworth filter inside the defined velocity cone.
Apply outside cone If selected, apply the Butterworth filter outside the defined velocity cone.
Low-cut filtering: Specify low-cut Butterworth filter parameters.
Apply low-cut filter If checked, apply a Butterworth filter with a low-cut.
Low-cut corner frequency (Hz) Enter the low pass frequency of the Butterworth filter in Hertz.
Low-cut rollof rate (dB/oct) Enter the low pass rolloff rate in dB/Octave. Higher numbers give a steeper filter rolloff.
High-cut filtering: Specify high-cut Butterworth filter parameters
Apply high-cut filter If checked, apply a Butterworth filter with a low-cut.
High-cut corner frequency (Hz) Enter the high pass frequency of the Butterworth filter in Hertz.
High-cut rollof rate (dB/oct) Enter the high pass rolloff rate in dB/Octave. Higher numbers give a steeper filter rolloff.
Usage:
The Velocity-Guided Noise Rejection Filter step is a 2D pre-stack single channel filter for attenuation of noise with a particular apparent velocity in the offset-time domain.
The input traces of each record are time-shifted according to their offsets, and then a low-cut spatial frequency filter is applied. This will attenuate much of the noise energy at that particular apparent velocity. The time-shifted data is then restored.
Input Links:
2) 2D Seismic pre-stack data records ordered by offset (mandatory).
Output Links:
2) 2D Seismic pre-stack data records ordered by offset (mandatory).
Reference:
-
Example Flowchart:
Step Parameter Dialog: