Nonlinear inversion for effective stress sensitive parameter using observed seismic data

Huaizhen Chen, Junxiao Li and Kris Innanen


Estimation of effective stress has become an important task in reservoir characterization and can guide the selection of fracturing area in unconventional hydrocarbon reservoirs. Based on Gassmann’s fluid substitution model, we propose a workflow of employing observed seismic data to implement nonlinear inversion for dry rock moduli, fluid factor and stress-sensitive parameter. We first make an approximation of fluid substitution equation, in which we replace the porosity term with a stress-sensitive parameter. Using stiffness parameters related to the stress-sensitive parameter, we derive a linearized reflection coefficient as a function of reflectivity of stress-sensitive parameter, and we also transfer the reflection coefficient to elastic impedance (EI). The proposed workflow involves estimating EI datasets from seismic data stacked over different ranges of incidence angle and utilizing the estimated EI to implement the inversion for the stress-sensitive parameter. We stress that a model-based least-squares inversion algorithm is used to implement the estimation of EI, and a nonlinear inversion approach is employed to estimate the unknown variables from the estimated EI, which is implemented as a four-step inversion. Synthetic data generated using Zoeppritz equation are utilized to verify the stability of the proposed approach. A test on real data set acquired over a gas-bearing reservoir reveals that the propose workflow appears to preserve as a useful tool to provide reliable results for fluid identification and stress prediction.

View full article as PDF (1.05 Mb)