DETERMINING THE BENEFITS OF SEISMIC HISTORY MATCHING
FOR FRAC
TURED RESERVOIRS
A joint project is proposed between Colorado School of Mines and Heriot-Watt University, in which the benefits of using conventional and anisotropic seismic attributes in a semi-automated seismic history match are investigated. The work builds on previous research by both universities in the area of seismic history matching and anisotropy analysis. The proposed investigation will focus in particular on quantifying the exact information content of such data in the context of permanent installations and/or frequently acquired 4D seismic data.
Seismic history matching is a semi-automatic procedure for matching both production and 4D seismic data to simulations from the reservoir model. By combining the excellent areal resolution of the ‘softer’ seismic data with the more areally sparse but ‘harder’ well production and pressure data, it is anticipated that the parameters of the reservoir model can be more accurately estimated. The procedure has possible benefits in a reduced uncertainty for forward predictions from reservoir simulation and hence better economic forecasts. A number of such seismic history matching schemes have been developed by various groups (for example, Stephen et al. 2004; Cominelli et al. 2002; Favergik et al. 2001), but all make use of PP data with isotropic seismic attributes. The aim of this project is to diversify the seismic attributes input to this method, and in particular to work with the additional possibility of anisotropy data. Anisotropy attributes have shown promise in the past to help define fractures (Davis 2004). Such data however require appropriate processing and interpretation, adequate cross-equalization, and bring their own set of problems. Although many of the difficulties in analyzing data for anisotropy are now fairly well understood, the advantages and disadvantages of using such attributes against the backdrop of a larger possible seismic error has not yet been fully evaluated, particularly in the context of a quantitative 4D analysis technique like seismic history matching.
Specifically, the aim of this project is to determine exactly what information content can be extracted from a joint PP isotropy and anisotropy seismic history match. The focus will be on:
1) permanent installations for recording and measuring various seismic attributes
2) frequently repeated data – the value of doing this, and what frequency of shooting is required to be beneficial;
3) the overall value of the seismic information – in terms of use of the final model for a forward economic forecast.
Both collaborating universities will participate equally, and lend their own specialized skills and experience to the project.
CSM role: the work of CSM builds on previous experience in the area of processing and analysis of seismic data for reservoir characterization. CSM will perform the acquisition and processing of the data donated to this project (Yamamoto, Fanchi and Davis, 2004; Davis 2004), and will be involved in quantification of the uncertainty in the isotropic and anisotropy attributes. The rock & fluid physics and interpretation components will be jointly considered by both universities.
HWU role: HWU will make use of their recently developed SHM workflow analysis, and previous experience of 4D seismic analysis and qualitatively matching of the seismic for reservoir model optimization (Stephen et al. 2004). They will perform the actual seismic history match and interact with CSM on how to incorporate the seismic data into the workflow.
There are two options for the project dataset:
1) Company dataset - a sponsoring company provides project participants with full access to an appropriate seismic dataset. The dataset could be either marine OBS or land, but should be provided along with full supporting non-seismic data. This should include a reservoir model, and full production and pressure data. All must be freely available at the start of the project. The ideal dataset should not be too complex structurally, and have a sufficient number of producing and injecting wells to adequately constrain the model parameters. It would be of advantage if the reservoir did not require additional understanding of geomechanical phenomena.
2) CSM dataset - CSM RCP will provide the dataset from its archives of existing data. Access to supporting non-seismic data will be provided and either a reservoir model built or an existing model refined. Weyburn might be a good choice in this context because one of the conclusions of Yamamoto, Fanchi and Davis (2004) that the reservoir performance should be matched using a dual porosity simulator, so it would be possible to apply HWU’s workflow and prepare a more sophisticated flow model using available data.
The project will be supported by four senior staff members, two from each university. In addition, a PhD student will be recruited for both CSM and HWU, supported by university staff on both sides. It is anticipated that both students and staff will spend some time at each other’s universities. Students will spend at least 3 months per year at the collaborating university. PhD degrees will be awarded separately from the each respective university. The following constitute the project team:
· Prof. Tom Davis, Department of Geology and Geophysics
· Prof. John Fanchi, Department of Petroleum Engineering
· CSM PhD student
· Prof. Colin MacBeth, Professor of Reservoir Geophysics, Institute of Petroleum Engineering.
· Dr. Karl Stephen, Senior Research Associate in Reservoir Engineering, Institute of Petroleum Engineering
· HWU PhD student
The duration of the project is three years, with the main costs being the tuition fees and maintenance of the two PhD students. The project seeks Industry sponsors at a total cost of £406k. The anticipated start date for the project is June 2005.
Staff cost plus overheads |
|
£120,000
|
|
2 PhD students maintenance + tuition fees |
|
£180,000
|
|
Benchfees |
|
£30,000
|
Hardware and software
|
|
£20,000
|
Travel and expenses for project personnel |
|
£56,000
|
|
TOTAL |
|
£406,000
|
Al-Naamani, A., and MacBeth, C., 2003. Constraining the reservoir model using PP
and PS data, First Break, 20, 235-246.
Cominelli, A., Seymour, R., Stradiotti, A., and Waggoner, J., 2002. Integrating time
-lapse seismic data in the history match of a gas-condensate reservoir.
Extended Abstracts, EAGE Conference, Florence, Italy.
Davis, 2004.
IEA GHG Weyburn CO2 Monitoring and Storage Project
Summary Report 2000-2004: From the Proceedings of the 7th International
Conference on Greenhouse Gas Control Technologies, Sept. 5-(th, Vancouver,
Canada, eds. M. Wilson and M. Monea. Favergik, K., Lygren, M., Valen, T.,
Hetlelid, A., Berge, G., Dahl, T., Sonneland, L., Lie, H., and Magnus, I., 2001. A
method for performing history matching of reservoir flow models using 4D seismic,
Expanded Abstracts, SEG Annual Meeting, San Antonio, Texas.
Stephen, K., Soldo, J., MacBeth, C., and Christie, M., 2004. Practical dynamic
updating of reservoir models using frequently acquire 4D seismic data. Extended
Abstracts, EAGE, Paris, France.
Yamamoto, H., Fanchi, J.R., and Davis, T.L., 2004, Integration of time-lapse seismic
data into a flow model study of CO2 injection into the Weyburn field, paper SPE