Summary
Traditional deterministic seismic inversion methods usually fail to depict fine reservoirs, due to the limited bandwidth, while stochastic seismic inversion (SSI) methods can offer us some multiple equiprobable models for further reservoir prediction. The basic idea of SSI is stochastically simulating non-well data basing on well data with a certain degree of randomness. The stochastic simulated data can just indirectly reflect the uneven distribution of reservoir characteristics subsurface. The easiest geostatistical simulation algorithm is the sequential Gaussian simulation (SGS). SGS combines the simulation and Kriging (well data conditioning) steps elegantly. In this abstract we present a different implementation procedure of SGS in stochastic seismic inversion. Corresponding numerical examples are carried out to check the effectiveness of the new method.
