Tomografia CRP

Author(s)

Gustavo Barroso Dias Ignácio

• Document type: Doctoral Thesis
• Press: Campinas, SP.
• Institution: Universidade Estadual de Campinas (UNICAMP). Faculdade de Engenharia Mecânica
• Defense date: 2021-01-26

Examining board members

Martin Tygel; Armando Lopes Farias; Jorge Henrique Faccipieri Junior; Rodrigo Bloot; Tiago Antonio Alves Coimbra

• Advisor: Martin Tygel

Abstract

Subsurface information extracted from seismic reflection data play a crucial role for exploration and production of oil and gas reservoirs. In this way, imaging and inversion techniques that provide such information are always in demand. Because of their importance, seismic velocities have been the main parameters for inversion. Because of that, the term velocity model building is very popular in the seismic literature. In fact, this terminology is referred to the inversion, not only of velocities, but also other model parameters that describe the geology of interest. Reliable and meaningful velocity models are essential to seismic imaging methods, most particularly migration. In fact, scientific and technological advances in migration come hand-in-hand with corresponding advances in velocity model building. The main focus of this thesis is on tomographic velocity model building. This thesis proposes a new tomography method, named CRP tomography. Based on the stereotomographic method from which it has several similarities, CRP tomography has, however, significant differences on the number and nature of the parameters involved in the inversion process. As opposed to stereotomography, for which the observed parameters are individually picked or extracted from the input data, CRP tomography makes use of parameter gathers, called CRP gathers, picked/estimated from the input data. More specifically, each CRP gather consists of source-receiver pairs within the input data for which the primary-reflection rays for a certain interface have the same (common) reflection point. In the same way as in the stereographic method, a few individual points are actually picked in CRP tomography. Referred to as reference points, these points are extended to corresponding CRP gathers extracted from the input data by means of coherence-analysis estimations performed on suitably defined parametric traveltimes. In fact, CRP tomography bears its name from such traveltimes, also referred to CRP traveltimes. This thesis shows how the addition of more information related to the same, previously unknown, point in depth, helps to constrain different kind of velocity models. All information related to each common-reflection-point are considered simultaneously to improve the localization of the particular related model common-depth-point. Practical tests that illustrates this thesis show that the use of CRP information acts as a natural constraint that is incorporated to the tomographic inverse problem. The CRP gathers allow more parts of the velocity model to be covered because more pairs of rays are traced and used bringing more internal information to the inversion process. Encouraging results provided by quite a few synthetic-data tests confirms those good expectations. As such, CRP tomography is seen to have a good the potential of being a reliable technique for velocity model building (AU)

• FAPESP's process: 16/19115-0 - 2D and 3D tomographic methods based on CRS parameters.
• Grantee: Gustavo Barroso Dias Ignácio
• Support Opportunities: Scholarships in Brazil - Doctorate