Elastic time-reverse imaging and transmission tomography for microseismic and DAS VSP data

Wavefield migration and tomography are considered to be state-of-the-art methodologies used for subsurface geological characterization. Seismic tomography produces accurate velocity models that commonly serve as input into seismic migration algorithms that produce high-quality passive-source (e.g., microseismic) images or structural images of geological interfaces constructed using controlled-source energy (e.g., vibroseis truck or dynamite). Most existing wavefield migration and tomography techniques employed in the oil and gas industry are well-developed under the acoustic assumption. One of the main shortcomings of this assumption is that conventional acoustic imaging algorithms generally use single-component P-wave data and thus do not account for multicomponent elastic (P- and S-mode) data that can provide additional subsurface information such as fracture distributions and elastic properties. To account for more accurate wave physics in passive and active seismic scenarios, I propose a suite of novel full-wavefield methods for imaging and multiparameter (i.e., P- and S-wave) model estimation in elastic media. Passive-style image-domain elastic tomography operates with multicomponent P- and S-wave first-arrival waveforms of a microseismic event and optimizes the background velocity model by improving the quality of source images constructed by a procedure called time-reverse imaging (TRI). To formulate a robust image-domain inversion framework, I develop a 3D extended imaging condition for surface-recorded microseismic data based on the correlation of individual P- and S-wavefield energy as well as the energy norm. The proposed PS energy imaging condition not only effectively locates microseismic events for complex isotropic/anisotropic models but also provides useful information about P- and S-wave velocity model as well as anisotropy parameter [ε,δ,γ] accuracy. Based on the kinetic energy term of the PS energy imaging condition, I propose an image-domain elastic wavefield tomography framework to build plausible P- and S-wave velocity models that improve the quality of microseismic event images. I present synthetic numerical experiments to demonstrate that the estimated model parameters result in enhanced source images, which greatly reduce event mispositioning errors. Finally, I apply the developed image-domain elastic inversion method on an active-source distributed acoustic sensing 3D vertical seismic profiling data set acquired in the North Slope of Alaska to investigate potential methane gas hydrate reservoirs. I exploit source-receiver reciprocity to create an acquisition configuration that resembles passive-seismic surface monitoring scenarios. I first validate the accuracy of the inverted elastic velocity models using a TRI-based source location analysis. Next, I construct numerous 3D structural images of the area of interest through elastic reverse time migration (RTM). The elastic RTM results exhibit coherent reflectivity associated with a complex near-surface ice-bearing permafrost zone, as well as two gas hydrate reservoirs that satisfactorily match the existing log data in well-ties due to the improved velocity model estimates.

Name thesis:degree_name

Doctor of Philosophy (Ph.D.)

Level thesis:degree_level

Doctoral

Discipline thesis:degree_discipline

Geophysics

Grantor dc:publisher

Colorado School of Mines. Arthur Lakes Library

Year dc:date.issued

2022

Author dc:creator

• Oren, Can

Advisor dc:contributor.advisor

• Shragge, Jeffrey

Committee members dc:contributor.committeemember

• Sava, Paul C.
• Bozdag, Ebru
• Ganesh, Mahadevan
• Walton, Gabriel

• computational seismology
• distributed acoustic sensing
• elastic wave propagation
• microseismic
• seismic imaging
• seismic tomography

Statement dc:rights

• Copyright of the original work is retained by the author.

Language dc:language.iso

eng, English

Identifier

T 9297

OAI identifier oai:identifier

oai:repository.mines.edu:11124/15380