Pore pressure prediction in orthotropic medium based on rock physics modeling of shale gas

Ting Lei,Xingyao Yin,Zhaoyun Zong

DOI: https://doi.org/10.1016/j.jngse.2019.103091

IF: 5.285

2020-02-01

Journal of Natural Gas Science and Engineering

Abstract:Pore pressure prediction is critical to reducing drilling costs and exploring hydrocarbons. Current seismic approaches for pore pressure prediction mainly initially establish the relation between vertical effective stress and physical parameters (velocity, resistivity or porosity), and then predict pore pressures with the known effective stress theory. The commonly used Eaton's equation requires the estimation of normal compaction trend (NCT), which cannot be measured but only be inferred. However, the conventional NCT is a curve which is able to describe the normal trend but is relatively rough, limiting the accuracy of pore pressure prediction. In this study, the shale rock physics modeling in orthotropic (OA) medium is designed to equate the NCT. The modeled and the measured P-wave velocity match well in normal pore pressure segments, but they differ in abnormal pore pressure segments. The modeled NCT is utilized to predict pore pressures, and the comparison with the measured pore pressures proves that the accuracy can be improved. Then P-wave velocity is inverted from azimuthal seismic data based on the reflectivity approximation of OA medium, and the pore pressure of the research area is predicted using the modeled P-wave velocity (the target play is a shale gas field in Sichuan basin of China). The prediction result is consistent with the characteristics of overpressure mechanism of hydrocarbon generation, indicating the effectiveness of rock physics modeling in pore pressure prediction.

energy & fuels,engineering, chemical

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to improve the accuracy of pore pressure prediction in shale gas reservoirs. Specifically, current seismic pore pressure prediction methods are mainly based on the relationship between vertical effective stress and physical parameters (such as velocity, resistivity or porosity), and use the known effective stress theory to predict pore pressure. Commonly used methods such as the Eaton equation require the estimation of the normal compaction trend (NCT), and NCT can only be inferred and not directly measured, which limits the accuracy of pore pressure prediction. This paper proposes a shale petrophysical modeling method based on orthotropic medium (OA) to more accurately simulate NCT, thereby improving the accuracy of pore pressure prediction.

### Main research contents
1. **Petrophysical modeling**:
- **Orthotropic medium (OA)**: Shale usually has strong vertical transverse isotropy (VTI) characteristics and can be assumed to be an OA medium due to the existence of bedding structures.
- **Steps of petrophysical modeling**:
1. **Clay block simulation**: Use the anisotropic self - consistent approximation (ani - SCA) and the anisotropic differential effective medium (ani - DEM) models to mix clay and kerogen to form clay blocks.
2. **Shale bedding simulation**: Calculate the elastic stiffness tensor of the rotated clay block in the observation coordinate system through Bond transformation, and use the Backus averaging theory to superimpose multiple rotated clay blocks to form layered shale.
3. **Establishment of dry rock skeleton**: Use the Hashin - Shtrikman - Hill (HSH) averaging model to mix brittle minerals (such as quartz, calcite, dolomite, feldspar and pyrite), and add brittle composites and pores to the shale matrix through the ani - DEM model. Fractures are added through the Hudson model.
4. **Shale saturation**: Use the Brown - Korringa equation to add fluids to complete petrophysical modeling.

2. **Pore pressure prediction**:
- **P - wave velocity inversion**: Invert the P - wave velocity from azimuthal seismic data using the reflectivity approximation method of OA medium.
- **Application of Eaton equation**: Use the simulated P - wave velocity as NCT, predict pore pressure through the Eaton equation, and compare it with the NCT calculated by the conventional method.

### Results and conclusions
- **Model verification**: The simulated P - wave velocity agrees well with the measured P - wave velocity in the normal pore pressure section, but there are differences in the abnormal pore pressure section, which may be due to the influence of abnormal pore pressure.
- **Pore pressure prediction**: The pore pressure predicted using the simulated NCT is more accurate than that predicted using the conventional method, verifying the effectiveness of petrophysical modeling in pore pressure prediction.

### Formulas
- **Terzaghi effective stress theory**:
\[
\sigma_v=\sigma_e + p
\]
where \(\sigma_v\) is the overburden pressure, \(\sigma_e\) is the vertical effective stress, and \(p\) is the pore pressure.

- **Eaton equation**:
\[
\Delta t_{\text{st}}=\left(\frac{\sigma_v-\sigma_n}{\sigma_v}\right)^c\Delta t_{\text{n}}
\]
where \(\Delta t_{\text{st}}\) is the measured acoustic wave travel time, \(\Delta t_{\text{n}}\) is the acoustic wave travel time under normal compaction conditions, \(\sigma_v\) is the overburden pressure, \(\sigma_n\) is the hydrostatic pressure, and \(c\) is a constant.

- **Normal compaction porosity**:
\[
\phi_{\text{n}}=\phi_0e^{-cZ}
\]
where \(\phi_{\text{n}}\) is the normal compaction porosity, \(\phi_0\) is the porosity at the ground or seabed, \(Z\) is the depth, and \(c\) is a constant.

Through these methods and models, this paper...