Loading...
Darcy flow coupled time-lapse gravity inversions for reservoir properties
Capriotti, Joseph
Capriotti, Joseph
Citations
Altmetric:
Advisor
Editor
Date
Date Issued
2020
Date Submitted
Collections
Research Projects
Organizational Units
Journal Issue
Embargo Expires
Abstract
The physical characteristics of a reservoir will always be important to understand, whether the goal is to maximize production of a hydrocarbon reservoir or predict the effects of a CO2 storage operation. Having a means to improve a reservoir model will help better determine these physical characteristics. Traditionally, a reservoir model is developed by adjusting parameters, such as permeability and porosity, that describe the reservoir until the predicted and observed production data match, a process known as history matching. Time-lapse geophysical measurements provide information about the distributions of fluids in the reservoir, supplementing the sparse information at the wells and have the potential to improve the reservoir models. However, common interpretation techniques of time-lapse geophysical data produce models that do not explicitly satisfy the governing equations of fluid flow. This problem is further compounded in gravity methods due to the static nature of the gravity field. The gravity response is sensitive to the whole volume of the subsurface and cannot be directly linked to a specific location in space, making the time-lapse gravity data difficult to correlate to reservoir parameters. The main motivation of my thesis is to develop Darcy-flow coupled time-lapse inversions for the time-lapse gravity method, in a way that is extensible to other time-lapse geophysical methods. Time-lapse gravity is directly sensitive to mass movement in the subsurface due to the exchange of fluids, making it a powerful method on its own. I first develop a method to solve for a permeability distribution from time-lapse gravity through a coupled inversion. After understanding the information which time-lapse gravity adds to the inversion for permeability, I next solve for both permeability and porosity simultaneously using similarity measures adapted from joint inversion. Because permeability is often anisotropic, I develop clustering inversion methodologies for anisotropic parameters by first investigating a simple anisotropic electrical conductivity structure, and then apply that methodology to the anisotropic permeability inversion. This work has the potential for significantly advancing the quantitative integration of non-seismic time-lapse geophysical data sets through coupled physics and by accounting for anisotropy.
Associated Publications
Rights
Copyright of the original work is retained by the author.