Development of Time Lapse VSP Integration Workflow: A Case Study at Farnsworth CO2-EOR Project

Abstract

This study aims to develop a 4D Vertical Seismic Profile (VSP) integration workflow to improve the prediction of subsurface stress changes. The selected study site is a 5-spot pattern within the ongoing CO2-EOR operations at the Farnsworth Field Unit FWU in Ochiltree County, Texas. The specific pattern has undergone extensive geological and geomechanical characterization through the acquisition of 3D seismic data, geophysical well logs, and core. This workflow constrains a numerical hydromechanical model by applying a penalty function formed between "modeled" versus "observed" time-lapse compressional and shear seismic velocity changes. Analyses of geophysical logs and ultra-sonic measurements on core exhibit measurable sensitivities to changes in both fluid saturation and mean effective stress. These data are used to develop a site-specific rock physics model and stress-velocity relationship, which inform the numerical models used to generate the "modeled" portion of the penalty function. The "observed" portion of the penalty function is provided by a novel elastic full-waveform inversion of the available 3D baseline and three monitor surveys to produce high-quality estimates of time-lapse compressional and shear seismic velocity changes. The modeling workflow accounts sequentially for fluid substitution and stress impacts. Hydrodynamic and geomechanical properties of the 3D coupled numerical model are estimated through geostatistical integration of well log and core data with 3D seismic inversion products. Changes in seismic velocities due to fluid substitution are computed using the Biot-Gassmann workflow and site-specific rock physics. Stress impacts on time-lapse seismic velocity changes are modeled from the effective stress output of the hydromechanical model and are initially based on the velocity versus effective stress relationship extracted from core mechanical testing. Based on the principle of superposition of seismic wavefields, seismic velocity changes attributed to fluid substitution and that due to changes in mean effective stress are treated as linearly additive. The modeled results are upscaled using Backus averaging to reconcile scale discrepancies between the modeled and measured datasets to formulate the penalty function. This manuscript presents the forward modeling process and concludes that for the base case, the seismic velocity changes due to mean effective stress dominates over the seismic velocity changes attributed to fluid substitution because of the extensive range of the pressure perturbations. Successful minimization of this penalty function calibrates the coupled hydrodynamic geomechanical numerical model and affirms the suitability of acoustic time-lapse measurements such as 4D-VSP for geomechanical calibration.

Quantitative interpretation of time‐lapse seismic data at Farnsworth field unit: Rock physics modeling, and calibration of simulated time‐lapse velocity responses

“…8 The porosity dependence of K dry, G dry, and K s, G s were investigated, and a Krief type characterization was performed, yielding a site-specific Krief coefficient. 39 An empirical model was followed relating the dry frame moduli to the solid matrix moduli and Biot coefficient. The resulting equation for the dry frame bulk and shear modulus is given as Krief expressed the Biot coefficient (∝) as a function of porosity…”

“…This study evaluated the matrix elastic properties by analyzing geophysical logs including dipole sonic, pulsed neutron, and magnetic resonance 8 . The porosity dependence of K dry , G dry , and $${K_s}$$ , $${G_s}\;$$were investigated, and a Krief type characterization was performed, yielding a site‐specific Krief coefficient 39 . An empirical model was followed relating the dry frame moduli to the solid matrix moduli and Biot coefficient.…”

Time‐lapse VSP integration and calibration of subsurface stress field utilizing machine learning approaches: A case study of the morrow B formation, FWU

“…12 Again, the seismic velocities are influenced by subsurface stress changes, resulting from fluid production or injection. 13 During CO 2 -enhanced oil recovery, the anticipated changes in the pore pressures leads to stress redistribution. 14,15 Hydromechanical models are employed to investigate the subsurface stress redistribution subsequent to changes in pore pressures.…”

“…Several publications on the calibration of a coupled hydro-mechanical model using attributes computed from a VSP dataset have been found. 13,[19][20][21] Attributes such as the seismic wave travel times, velocities, amplitudes, and seismic reflectivity have been utilized to update reservoir flow properties and monitor the preferential fluid movement. 4,5,11,22 For instance, In Salah, a well-known CCUS project in Algeria has CO 2 stripped from produced natural gas and reinjected into the Krebka formations' aquifer leg.…”

“…Shear and compressional seismic velocities are affected by changes in fluid saturation 12 . Again, the seismic velocities are influenced by subsurface stress changes, resulting from fluid production or injection 13 . During CO 2 ‐enhanced oil recovery, the anticipated changes in the pore pressures leads to stress redistribution 14,15 …”