I introduce the motivation for the proposal of inverting for the reflectivity model and the velocity (background model), to improve the inversion result for both parameters. The proposed method is joint inversion of reflectivity and background subsurface component.
PDF |
Source code

Chapter 2

I formally derive the JIRB method as a modification of conventional linearized waveform inversion.
PDF |
Source code

Chapter 3

I show the implementation of the random boundary condition for the RTM and WEMVA components of the JIRB operator.
PDF |
Source code

Chapter 4

I apply the method to the Sigsbee, 2D synthetic model.
PDF |
Source code

Chapter 5

I apply the method to a 3D ocean-bottom node dataset.
PDF |
Source code

Appendix A

I show here the problem of linearizing the image norm, for the initial attempt to set the method as a linear optimization problem.
PDF |
Source code

Appendix B

I review how preconditioning using B-splines is implemented in WEMVA.
PDF |
Source code

Copyright (c) 2020 Stanford Exploration Project. All rights reserved.