Bayesian Lithology-Fluid Prediction and Simulation from Pre-stack Seismic Data
by Prof. Henning Omre
Abstract: The objective is to determine the Lithology-Fluid classes along 1D
profiles through a reservoir target zone. A stationary Markov chain prior
model is used. The likelihood model entails strong spatial coupling
related to the convolutional model. An approximation of the likelihood
model provides a posterior model which is a non-stationary Markov chain.
The posterior model can be assessed by and exact and efficient recursive
sampling algorithm. The approach is evaluated on a synthetic 1D profile
inspired by a North Sea sand stone reservoir.
The talk will contain a short summary of status in pre-stack seismic
Bayesian inversion, and discuss particular problems related to
For More Information: Emma Lockwood tel: 8344-1617 email: email@example.com