Quantitative reservoir characterization integrating seismic data and geological scenario uncertainty
Abstract/Contents
- Abstract
- The main objective of this dissertation is to characterize reservoir models quantitatively using seismic data and geological information. Its key contribution is to develop a practical workflow to integrate seismic data and geological scenario uncertainty. First, to address the uncertainty of multiple geological scenarios, we estimate the likelihood of all available scenarios using given seismic data. Starting with the probability given by geologists, we can identify more likely scenarios and less likely ones by comparing the pattern similarity of seismic data. Then, we use these probabilities to sample the posterior PDF constrained in multiple geological scenarios. Identifying each geological scenario in metric space and estimating the probability of each scenario given particular data helps to quantify the geological scenario uncertainty. Secondly, combining multiple-points geostatistics and seismic data in Bayesian inversion, we have studied some geological scenarios and forward simulations for seismic data. Due to various practical issues such as the complexity of seismic data and the computational inefficiency, this is not yet well established, especially for actual 3-D field datasets. To go from generating thousands of prior models to sampling the posterior, a faster and more computationally efficient algorithm is necessary. Thus, this dissertation proposes a fast approximation algorithm for sampling the posterior distribution of the Earth models, while maintaining a range of uncertainty and practical applicability. Lastly, the proposed workflow has been applied in an actual reservoir. The field, still in the early stage, has limited well data, seismic data, and some geological observations. Accordingly, the proposed workflow can guide several processes, from selecting geological scenarios to suggesting a set of models for decision makers. The case study, applied in a turbidite reservoir in West Africa, demonstrates the quantitative seismic reservoir characterization constrained to geological scenarios. It contains a well log study, rock physics modeling, a forward simulation for generating seismic responses, and object-based prior modeling. As the result, we could pick some promising geological scenarios and its geological parameters from seismic data using distance-based pattern similarity. Next, based on the selected geological scenarios, Metropolis sampler using Adaptive Spatial Resampling (M-ASR) successfully sampled the posterior conditioned to all available data and geological scenario uncertainty.
Description
| Type of resource | text |
|---|---|
| Form | electronic; electronic resource; remote |
| Extent | 1 online resource. |
| Publication date | 2014 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Associated with | Jeong, Cheolkyun |
|---|---|
| Associated with | Stanford University, Department of Energy Resources Engineering. |
| Primary advisor | Mukerji, Tapan, 1965- |
| Thesis advisor | Mukerji, Tapan, 1965- |
| Thesis advisor | Caers, Jef |
| Thesis advisor | Mavko, Gary, 1949- |
| Advisor | Caers, Jef |
| Advisor | Mavko, Gary, 1949- |
Subjects
| Genre | Theses |
|---|
Bibliographic information
| Statement of responsibility | Cheolkyun Jeong. |
|---|---|
| Note | Submitted to the Department of Energy Resources Engineering. |
| Thesis | Thesis (Ph.D.)--Stanford University, 2014. |
| Location | electronic resource |
Access conditions
- Copyright
- © 2014 by Cheol Kyun Jeong
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...