Time-lapse Inverse Theory

by Musa Maharramov


Table of contents

Compaction in reservoir overburden can impact production facilities and lead to a significant risk of well-bore failures. Prevalent practice of time-lapse seismic processing of 4D data above compacting reservoirs relies on picking time displacements and converting them into estimated velocity changes and subsurface deformation. This approach relies on prior data equalization, and requires a significant amount of manual interpretation and quality control. In this thesis I develop methods for automatic detection of production-induced subsurface velocity changes from seismic data, and computational techniques of subsurface characterization from measurable surface deformation.
In the first part of my work I describe a time-lapse inversion technique based on a simultaneous regularized full-waveform inversion of multiple surveys. I provide a theoretical foundation of the proposed method, and analyze its sensitivity to a realistic 1-2&#37 velocity deformation in the overburden. The method is applied in a study of overburden dilation above the Gulf of Mexico Genesis field and achieves a stable recovery of &quotblocky&quot negative velocity anomalies above compacting reservoirs. I propose a multi-scale extension of the method for recovering both long and short-wavelength velocity changes.
In the second part I describe a geomechanical model of overburden deformation in response to fluid extraction or injection. I provide a method for inverting pore pressure changes from noisy and sparse measurements of surface deformation. The method is applied to estimating the efficiency of Cyclic Steam Stimulation (CSS) of a heavy oil reservoir. Inverted subsurface pore pressure changes indicate a significant heterogeneity of the propagating steam fronts. The method is extended to a study of sharp contrasts in reservoir attributes by using a new splitting algorithm for solving large-scale constrained regularized optimization problems.