Joint Inversion of Time-Lapse Seismic Data
- Univ. of North Dakota, Grand Forks, ND (United States)
The Energy & Environmental Research Center (EERC) conducted the Joint Inversion of Time-Lapse Seismic Data project in which two joint inversion modeling and monitoring workflows were developed to address and resolve shortcomings of existing inversion technology and time-lapse amplitude difference interpretation. The first workflow is based on the seismic wave-equation based (WEB) amplitude variation with offset (AVO) inversion which was performed in collaboration with project partner Delft Inversion. The nonlinear WEB-AVO inversion solves the full elastic wave-equation for the properties as well as the total wave-field in the object domain. The Joint Impedance and Facies Inversion (Ji-Fi) developed by project partner Ikon Science is the other technique applied in this project. This technique is based on Bayesian principles, geologic facies as priors, and powerful image-processing techniques to invert seismic data iteratively for impedances given facies and inverting the impedances for facies. Both techniques were successfully applied to an existing time-lapse seismic data set. The WEB-AVO workflow is based on a new feature of the technology specifically developed for this project: the simultaneous joint inversion of the monitor and baseline. One of its main advantages is the robustness against the noise associated with the “no repeatability” of seismic surveys. This development of the WEB-AVO inversion is part of the process to upgrade the technique from a technology readiness level (TRL) 3 to TRL4. The estimated time-lapse changes of WEB-AVO compressibility and shear compliance were assessed, and it was concluded that shear compliance is a good indicator of the pressure effect due to the CO2 enhanced oil recovery (EOR) activities in the study area. This ability to separate the effect of pressure from CO2 saturation can be used to better assess the location of CO2 within the target reservoir. As part of the Ji-Fi workflow, the facies and their probability distributions corresponding to the highly heterogeneous target reservoir and its fluid conditions due to the CO2 EOR activities were successfully separated in the seismic elastic space. This separation provided favorable conditions for the application of the Ji-Fi method. Reliable acoustic impedances and, most probably, litho-fluid facies of the target reservoir were obtained from the application of the Ji-Fi method separately to the baseline and monitor surveys. Information about the probability of the presence of litho-fluid facies is paramount to reducing uncertainty in forecasting CO2 saturation changes within the target reservoir. WEB-AVO compressibility data were integrated into the reservoir simulation model by employing two widely accepted compressibility–porosity correlations for sandstones. Flow mechanism analysis and engineering calculations showed that pore and fluid compressibility can affect the pressure response in the reservoir. A closed-loop workflow has been developed in this study to integrate the seismic inversion data into an existing reservoir simulation model and update the model with different pore compressibility distribution scenarios. The comparison of simulation results and field observations confirms the estimated time-lapse seismic response for compressibility distribution and saturation in the Bell Creek Field.
- Research Organization:
- Univ. of North Dakota, Grand Forks, ND (United States)
- Sponsoring Organization:
- USDOE
- DOE Contract Number:
- FE0031540
- OSTI ID:
- 1619054
- Report Number(s):
- DOE-EERC-FE0031540
- Country of Publication:
- United States
- Language:
- English
Similar Records
Small Scale Field Test Demonstrating CO2 Sequestration In Arbuckle Saline Aquifer And By CO2-Eor At Wellington Field, Sumner County, Kansas
Robust Carbon Dioxide Plume Imaging Using Joint Tomographic Inversion of Seismic Onset Time and Distributed Pressure and Temperature Measurements (Final Report)