Inconsistency of Rock physics model predictions \& Anisotropic time-lapse tomographic results

Nicolas Mastio and Pierre Thore and Marianne Conin and Guillaume Caumon. ( 2019 )
in: 2019 Ring Meeting, ASGA

Abstract

In the petroleum industry, time-lapse (4D) studies are commonly used for reservoir monitoring but are also useful to perform risk assessment in the overburden (well shearing, cap rock integrity ...). Although complex anisotropic velocity changes are predicted in the overburden by geomechanical studies, the classical time-lapse inversion workflow only deals with vertical velocity changes. To retrieve the geomechanically induced anisotropy of the 4D signal in a real case study, we propose to use a reflection traveltime tomography method coupled with a warping algorithm. We test the workflow on a real dataset. We first built synthetic data with the predicted velocity found in the literature (vertical P-wave velocity decrease in the cap rock and an isotropic P-wave velocity change in the reservoir) to check which ray angular coverage is necessary to retrieve the targeted anisotropy. We show that the retrieved anisotropies depend on offset range and that the acquisition of the real case study is suitable to perform the tomographic inversion. Applying the tomography to the real data, we find that the velocity change in the overburden is unexpectedly mainly horizontal or oriented in the 45 degree direction. Our interpretation of the orientation of the anisotropy is that the 4D signal in the overburden is mainly due to preexisting joints and faults or can be the evidence of rock shear failure during production.

Download / Links

BibTeX Reference

@inproceedings{MastioRM2019,
 abstract = { In the petroleum industry, time-lapse (4D) studies are commonly used for reservoir monitoring but are also useful to perform risk assessment in the overburden (well shearing, cap rock integrity ...). Although complex anisotropic velocity changes are predicted in the overburden by geomechanical studies, the classical time-lapse inversion workflow only deals with vertical velocity changes. To retrieve the geomechanically induced anisotropy of the 4D signal in a real case study, we propose to use a reflection traveltime tomography method coupled with a warping algorithm. We test the workflow on a real dataset. We first built synthetic data with the predicted velocity found in the literature (vertical P-wave velocity decrease in the cap rock and an isotropic P-wave velocity change in the reservoir) to check which ray angular coverage is necessary to retrieve the targeted anisotropy. We show that the retrieved anisotropies depend on offset range and that the acquisition of the real case study is suitable to perform the tomographic inversion. Applying the tomography to the real data, we find that the velocity change in the overburden is unexpectedly mainly horizontal or oriented in the 45 degree direction. Our interpretation of the orientation of the anisotropy is that the 4D signal in the overburden is mainly due to preexisting joints and faults or can be the evidence of rock shear failure during production. },
 author = { Mastio, Nicolas AND Thore, Pierre AND Conin, Marianne AND Caumon, Guillaume },
 booktitle = { 2019 Ring Meeting },
 publisher = { ASGA },
 title = { Inconsistency of Rock physics model predictions \& Anisotropic time-lapse tomographic results },
 year = { 2019 }
}