0% found this document useful (0 votes)
268 views13 pages

High-Frequency Enhancement of Seismic Data by Reconvolution

This document describes a method for high-frequency enhancement of seismic data through reconvolution. The method involves (1) calculating the first derivative of seismic traces, (2) identifying zero-crossings to create a sparse-spike series, and (3) convolving the spike series with a higher-frequency wavelet. This process increases the high-frequency content of the seismic data while preserving amplitudes at spikes. The document provides examples of applying this method using Hampson-Russell software.

Uploaded by

Mark Ralph
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
268 views13 pages

High-Frequency Enhancement of Seismic Data by Reconvolution

This document describes a method for high-frequency enhancement of seismic data through reconvolution. The method involves (1) calculating the first derivative of seismic traces, (2) identifying zero-crossings to create a sparse-spike series, and (3) convolving the spike series with a higher-frequency wavelet. This process increases the high-frequency content of the seismic data while preserving amplitudes at spikes. The document provides examples of applying this method using Hampson-Russell software.

Uploaded by

Mark Ralph
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 13

High-Frequency Enhancement of Seismic Data by Reconvolution

Kevin Gerlitz

This method comes from a recent paper at the PETEX 2004 conference called “Cosmetic
Enhancement of Seismic Data By Loop Reconvolution” by P. Young and A. Wild. This paper was
brought to my attention by John Logel of Anadarko, Calgary.
The basic idea is that you first create a sparse-spike representation of the seismic data and then
convolve with a high-frequency wavelet.
The algorithm is as follows:
1) calculating the first derivative of the seismic trace,
2) identify the zero-crossings (which correlate to the maxima & minima on the seismic) and
3) convolve the spikes with a higher-frequency wavelet.

This process is illustrated graphically in the following slides…


1) The Seismic Trace and the Calculated First Derivative
2) Search the derivative trace for zero-crossings and replace with the coincident seismic amplitude –
everything else in the trace is zero.
A comparison of the original seismic with the sparse spike series
3) Convolve the sparse-spike series with a high-frequency wavelet. In this example, I applied a large
bandwidth bandpass filter.
The following slides illustrate how to do this using the Hampson-Russell Software package.
Resample the input seismic to a higher sample rate. In this example, I resampled the 2 ms dataset to 0.5
ms…
Calculate the derivative of the resampled seismic. I used the Process > Utility > Seismic Attribute
process to do this…
Double-click on the script to open up
the WordPad editor. Save it (Ctrl-A,
Ctrl-C) to clipboard and paste (Ctrl-V)
Now apply the Trace Maths script to the into the software.
two input volumes… /* Frequency Enhancement */

/* this script requires that the two input volumes */


/* i.e., the resampled seismic and the derivative, */
/* are named in and deriv, respectively. */

/* initialize the output volume */


out = in * 0.0;

/* get the number of samples of the traces */


length = numsamples( in );

i = 0;
while( i < length - 1 )
/* NB: testing for the (i+1)th sample so go to (length-1) */
{
/* is there a zero-crossing? */
if( sign( deriv[ i ] ) != sign( deriv[ i + 1 ] ) )
out[ i + 1 ] = in[ i + 1 ];

i = i + 1;
}

/* output the scaled spike volume */


out;

This script requires that the two input


volumes be named in and deriv
The output is a spiked time series. Filter it to replicate a wavelet convolution…
The output after filtering with a 5-10-90-120 bandpass filter.
The original for comparison
The original seismic amplitude spectrum The new enhanced seismic amplitude spectrum

You might also like