Matching of Surface seismic, VSP and Crosswell seismic data with the reflectivities stochastic inversion

摘要

Matching of the surface seismic, VSP and crosswell seismic datasets is the basis for integration application of different reservoir seismic data. Reflection coefficients have been selected as the bridge that matching for different seismic data, which is the linear convolution with the seismic wavelet respectively. A Bayesian framework is presented to estimate the reflection coefficient that combined with three different seismic dataset. The posteriori PDF of the reflectivities have been sampled with the popular Markov Chain Monte Carlo (MCMC) method. Comparison to the inversion result of the single seismic data, the stochastic inversion reflectivity realizations are matched the reflection characteristics of the surface seismic, VSP and crosswell seismic datasets simultaneously very well. This result is effective to explain the difference in reservoir seismic data at the same spatial location. The field data example shows that this stochastic inversion method provides a bridge for the matching of reservoir seismic data, which is the basis of the reservoir parameters simulation with different reservoir geophysical data.