Simulation of acoustic impedance images by stochastic inversion of post-stack seismic reflection amplitudes and well data

摘要

Here we present an approach for simulation of 3D images of acoustic impedance conditional both to post-stack seismic reflection amplitudes and well impedance data. The method uses the following steps: (1) we calculate an initial deterministic image of impedances and reflection coefficients; (2) we apply a modified version of the sequential simulation and co-simulation algorithm for generating the impedance images conditioned to the reflection coefficients previously calculated and the impedance observed in the wells; and (3) we validate the results, particularly with respect to the correlation between the observed seismic cubes and the synthetic seismic cubes obtained by convolution of the simulated images of impedance. The advantages of this workflow are (1) efficiency - it is necessary to run only one simulation for each output image; (2) mapping of uncertainty - the set of equiprobable output images is conditional to both the impedance well data and the reflection coefficients of the deterministic image; and (3) compliance - the method complies with the characteristics of a simulation, namely, the declustered histogram of the well data, the variogram, and the reproduction of well data in blocks close to well locations. The approach was tested for modeling the impedance of a selected region of the Stratton gas reservoir.