Gas Exploration Software for Reducing Uncertainty in Gas Concentration Estimates
APPLICATIONS OF TECHNOLOGY:
Improve estimation of reservoir parameters and quantify uncertainty in the estimation when exploring for gas and oil deposits using geophysical data
• Provides extensive uncertainty information to reduce the chances of drilling in uneconomical locations
• Exists in two versions, C++ or Fortran 90
• Provides clear interfaces for users to plug in their own codes for rock-physics models and for forward simulation of seismic AVA and CSEM data
• General and flexible, allowing for a wide range of prior information to be included into the estimation
• MCMC sampling results can be read by R, S-PLUS and SAS software for analysis
Estimating reservoir parameters for gas exploration from geophysical data is subject to a large degree of uncertainty. Seismic imaging techniques, such as seismic amplitude versus angle (AVA) analysis, can provide good information about the physical location and porosity of potential gas-bearing sands but cannot discriminate between economical and uneconomical gas concentrations. Using seismic AVA data alone, even with high resolution, it is difficult to distinguish high or low gas concentration in deep layers because seismic properties are not sensitive to variations in gas concentration. With the inclusion of controlled-source electromagnetics (CSEM) data, uncertainty in gas saturation estimation decreases, and the ability to identify high gas concentration in deep layers is enhanced.
Jinsong Chen of Lawrence Berkeley National Laboratory has developed software that helps to significantly refine estimates of gas or oil saturation by providing sufficient information on uncertainty, in the form of a clearly defined percentage range. Whereas traditional methods give only estimates of unknown parameters, Chen’s software can provide both estimates of the unknown parameters and their associated uncertainty information. This greatly enhances the possibility of identifying high gas concentration in deep layers and reduces the risk of uneconomical drilling.
Chen’s software provides several Markov chain Monte Carlo (MCMC) sampling methods to explore the joint posterior distribution function within the Bayesian framework. These can be flexibly used for inverting various types of unknown parameters. The method is applied to both synthetic and field data, and can be extended for joint inversion of 2D or even 3D CSEM data.
The software is highly flexible and exists in both C++ and Fortran 90 versions. It can also be easily adapted to incorporate researchers’ own forward simulation codes of seismic and CSEM data and rock-physics models.
Type of Offer:
« More Environment Patents« More Ecology Patents