Tomographic full waveform inversion
Abstract/Contents
- Abstract
- Full waveform inversion (FWI) is an extremely powerful yet challenging procedure for inverting seismic data in exploration seismology. Its great potential lies in its ability to produce high resolution subsurface models of the earth by directly inverting the recorded waveform using the full wave equation. This direct matching approach allows FWI to simultaneously invert all components of the data resulting in the least possible ambiguity and most accurate results that can be extracted out of the seismic data. However, the high accuracy of FWI comes with high sensitivity to errors in the starting velocity model and local minima issues. In other words, the same reasons that give FWI its great features also give it its most challenging requirement: a very accurate starting model. Failure to satisfy this requirement results in cycle skipping. This strict initial model requirement is more pronounced when inverting reflection data because of the increased complexity and depth ambiguity. As a result, most of the successful applications of FWI only use the early arrival data. The central goal of this thesis provides a new inversion procedure that retains all the advantages and benefits of FWI while avoiding its strict initial model requirement and cycle-skipping challenges. To achieve this goal, I modified FWI by combining its classical form with a modified form of wave-equation migration-velocity analysis (WEMVA). This combination manifests itself as an extension of the velocity model through virtual axes. I named the new method tomographic full waveform inversion (TFWI). By extending the velocity model with the appropriate axis, the modeling operator can match the observed data regardless of the accuracy of the starting model by using kinematic information from the extended axis even when cycle skipping occurs. We set up the inversion to extract all the necessary information from the virtual axes and smoothly collapse them back into the physical, nonextended form of the model. While it avoids the cycle-skipping problem, TFWI does come with its own challenges—namely, its high-computational costs and high number of iterations required. Throughout this thesis, several modifications, analysis, and solutions are provided to overcome these challenges and allow TFWI to be both robust and feasible. The final algorithm is applied to both synthetic and field data sets. In all examples, the initial model is very inaccurate, resulting in cycle skipping in all frequencies of the residual. The results show remarkable reconstructions of most features of the models for all scales where TFWI successfully inverts the kinematic and dynamic information of the data with outstanding robustness and accuracy.
Description
| Type of resource | text |
|---|---|
| Form | electronic; electronic resource; remote |
| Extent | 1 online resource. |
| Publication date | 2016 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Associated with | Almomin, Ali Ameen M |
|---|---|
| Associated with | Stanford University, Department of Geophysics. |
| Primary advisor | Biondi, Biondo, 1959- |
| Thesis advisor | Biondi, Biondo, 1959- |
| Thesis advisor | Claerbout, Jon F |
| Thesis advisor | Clapp, Robert G. (Robert Graham) |
| Thesis advisor | Dunham, Eric |
| Advisor | Claerbout, Jon F |
| Advisor | Clapp, Robert G. (Robert Graham) |
| Advisor | Dunham, Eric |
Subjects
| Genre | Theses |
|---|
Bibliographic information
| Statement of responsibility | Ali Ameen M Almomin. |
|---|---|
| Note | Submitted to the Department of Geophysics. |
| Thesis | Thesis (Ph.D.)--Stanford University, 2016. |
| Location | electronic resource |
Access conditions
- Copyright
- © 2016 by Ali Ameen M Almomin
Also listed in
Loading usage metrics...