Scribd
Upload
9 views

GEOP 320 Spring 2012 Lab Assignment # 2 Deconvolution

This document provides instructions for a lab assignment on seismic data deconvolution. The objective is to (1) enhance vertical resolution and (2) remove multiples from seismic data. Key steps include: selecting parameters for the autocorrelation window, prediction lag, operator length, and percent prewhitening; deconvolving the provided data set; examining autocorrelations for multiples; selecting deconvolution parameters; analyzing the amplitude spectrum; applying optional band-pass filtering; and assessing if deconvolution enhanced resolution. Suggestions for improving results are also requested.

Uploaded by

Mark Mao
Copyright
© Attribution Non-Commercial (BY-NC)
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
9 views

GEOP 320 Spring 2012 Lab Assignment # 2 Deconvolution

This document provides instructions for a lab assignment on seismic data deconvolution. The objective is to (1) enhance vertical resolution and (2) remove multiples from seismic data. Key steps include: selecting parameters for the autocorrelation window, prediction lag, operator length, and percent prewhitening; deconvolving the provided data set; examining autocorrelations for multiples; selecting deconvolution parameters; analyzing the amplitude spectrum; applying optional band-pass filtering; and assessing if deconvolution enhanced resolution. Suggestions for improving results are also requested.

Uploaded by

Mark Mao
Copyright
© Attribution Non-Commercial (BY-NC)
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/ 1

1

GEOP 320 Spring 2012 Lab assignment # 2 Deconvolution (Due date: 16/4/2012) Objective The objective of this lab assignment is to deconvolve your dataset. Introduction Deconvolution is used for two main purposes: (1) Spiking the data to enhance the vertical resolution. (2) Removing multiples from the data. These two processes can be done by selecting appropriate values for the following parameters: (1) autocorrelation window (w) (2) prediction lag () (3) operator length (n) (4) percent prewhitening () Exercises Observing the specific parameters of your data set data2012-gain-edit-flt.su: 1. Follow the steps in tutorial 3.3 of the Manual and save your deconvolved data set as data2012-gain-edit-flt-dec-bal.su. 2. Generate 3 plots similar to the ones in the tutorial. 3. Answer the following questions: a. Examine the autocorrelations for the existence of long-path multiples. Are there any? Justify your answer. b. Examine the autocorrelations for the existence of short-path multiples. Are there any? Justify your answer. c. Provide a table that shows your selections and justifications for , n, , and w that you used to spike the data. d. Take a look at the FFT amplitude spectrum of the deconvolved data using the command: suwind key=ep min=1 max=6 < data2012-gain-edit-flt-dec-bal.su | suspecfx | suxwigb i. You observe generally an equal distribution of amplitude with frequency. Why does the amplitude spectrum behave so? ii. There might be unwanted noise appearing in the data after deconvolution. In this case, it is convenient to apply a band-pass filter to get rid of this noise using the following procedure: Use the command sufilter to filter the boosted noise and keep the signal and save the filtered data set as data2012-gain-edit-flt-dec-bal-flt.su. Provide printouts of shots 1-6 after filtering. Provide a table consisting of your filter parameters and justify your selections. e. After these processes, did deconvolution enhance the resolution of your data? If not, why? f. What do you suggest to deconvolve the data and still get better results?

You might also like