(Ph.D Dissertation)

Xiang Xiao, University of Utah

Click here to view the HTML version of the thesis.
Click here to download the Microsoft PowerPoint presentation of the oral defense. (7.3M)
Click here to download the PDF file of the thesis. (7.8MB)
Click here to download the PostScript file of the thesis. (109MB)
Click here to download the Latex texts and the PostScript figures for the thesis. All the Latex texts and PS files are tared into a tar file xxiao_thesis.tar.gz (47MB)


Three methods are introduced for imaging the reflectivity distribution with natural VSP Green's functions: $SSP \rightarrow
VSP \rightarrow SWP$ interferometric datuming and imaging, local reverse time migration for subsalt reflector imaging, and local reverse time migration for salt flank imaging. The transmitted arrivals in the recorded VSP Green's function are back-projected from the well while in standard migration these transmitted arrivals are forward modeled. This means that the overburden velocity model is not needed for migration and so leads to a significant improvement in imaging. There are three main chapters in this dissertation.

Chapter 2. Surface seismic profile (SSP) data can be naturally redatumed below salt with the natural Green's functions obtained from a VSP shot gather. No overburden or salt velocity model is needed and the quality of the image only depends on the accuracy of the $v(x,y,z)$ around the well and the receiver aperture width. The subsalt image is more accurate than that provided by migrating the SSP or VSP data alone. The numerical images show that they inherit the illumination aperture of SSP data, which is usually wider than the VSP data.

Chapter 3. A VSP imaging method that only requires a target-oriented local velocity model around the well is presented in the form of a Kirchhoff integral. Here, we call this method `local reverse time migration'. The complex overburden and salt body geometry are not needed, and the source-side statics are automatically accounted for. All kinematic and dynamic effects, including anisotropy, absorption and all other unknown/undefined rock effects outside of this local velocity model are automatically accounted for. The numerical tests empirically validate this method by accurately imaging the sediments next to the vertical well.

Chapter 4. The local reverse time migration method is adapted for imaging salt flanks by transmitted P-to-S waves. There are two strategies: separation of $PP$ and $PS$ transmitted wavefields before acoustic backward projection and `separation while imaging'. Here, only a localized velocity model and the receiver wavefields are used to delineate the salt boundaries with transmitted P-to-S waves. Numerical tests on data associated with an elastic salt model and Gulf of Mexico VSP data demonstrate its superiority over the traditional migration method.