Browse Prior Art Database

A method for a QC of elastic behaviour and/or an incorrect velocity model in seismic data

IP.com Disclosure Number: IPCOM000243577D
Publication Date: 2015-Oct-02
Document File: 2 page(s) / 99K

Publishing Venue

The IP.com Prior Art Database

Abstract

After building a high resolution P-wave velocity model using an algorithm based on the acoustic wave equation (for example, from an acoustic FWI inversion), an important QC is to quantify the size of the elastic effects and also possible errors in the velocity model. Our proposed method does this by: 1) forming a scaled difference (the residual) of the scaled real and modelled shot gathers, 2) migrating these residual shot gathers to give an attribute section that quantifies the elastic effects and/or possible errors in the velocity model. QC of elastic effects is important in evaluating the quality of a given velocity model and also identifying areas that might need further improvement in the velocity model building process. The proposed QC allows both of these elements to be investigated together

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 2

Title:A method for a QC of elastic behaviour and/or an incorrect velocity model in seismic data

Abstract:

After building a high resolution P-wave velocity model using an algorithm based on the acoustic wave equation (for example, from an acoustic FWI inversion), an important QC is to quantify the size of the elastic effects and also possible errors in the velocity model. Our proposed method does this by: 1) forming a scaled difference (the residual) of the scaled real and modelled shot gathers, 2) migrating these residual shot gathers to give an attribute section that quantifies the elastic effects and/or possible errors in the velocity model.

Description:

The methodology for generating the QC can be generally defined as follows:


1) Forward model through the estimated velocity model using the acoustic wave equation.


2) Extract this forward modelled wavefield at the receiver locations to give a modelled shot gather.


3) Match the relative amplitudes of the real and modelled shot gathers in an offset, and time, variant manner. This could be done using something as simple as a trace by trace time varying AGC, or a time varying amplitude only matching filter.


4) Subtract the real and modelled matched shot gathers, giving rise to a residual shot gather.


5) Optionally, the original amplitude behaviour with offset and time of the real dataset can be re-imposed on this residual shot gather to give a better balance to the residual amplitudes.


6) Repeat steps 1)-5) for each shot gather, or a subset of the shot gathers, in the survey.


7) Migrate this set of residual shot gathers to form an image.

Even if we assume that our P-wave velocity mode...