Model-based Approach to Automatic 3D Seismic Horizon Correlation across Faults

Seismic data provide detailed information about subsurface structures. Reflection events visible in the seismic data are known as horizons, and indicate boundaries be- tween different rock layers. A fault is a surface along which one side of rock layers has moved relative to the other in a direction parallel to the surface. Faults are recognized in seismic data by discontinuities of horizons. Fault interpretation is an important but time-consuming task of seismic interpretations. It is least supported by automatic tools due to seismic data distortions near fault regions. In this paper, we present an automatic method for correlations of horizons across faults in 3d seismic data. As automating horizons correlations using only seismic data features is not feasible, we find optimal correlations by imposing geological constraints. We formulate the correlation task as a non-rigid continuous point matching between the two sides of the fault. A fault surface is identified and seismic features on both sides of the fault are gathered. One side of the fault serves as the floating image while the other side is the reference image. A fault displacement model is constructed by per- forming initial discrete match of some prominent regions. Then the simulated anneal- ing optimization technique is used to perform continuous point matching between the two sides according to seismic feature similarity and the fault model. A possible multi- resolution scheme for the simulated annealing optimization is discussed. The method was applied to real 3D seismic data, and has shown to produce geologically acceptable results.