Accounting for Seismic Trends in Stochastic Well Correlation

Abstract

AbstractStratigraphic well correlation is a critical step of basin and reservoir analysis and modeling workflows. In this paper, we propose an automatic stratigraphic well correlation method which is based on both borehole data and interwell information extracted from poststack seismic data to constrain stratigraphic well correlation. The presented stratigraphic well correlation method uses the Dynamic Time Warping algorithm. Global correlations are built by combining elementary correlation costs between stratigraphic units or markers identified along studying wells. Whereas various rules can be used to compute the correlation likelihood between well sections, a significant challenge is to compute the cost for an unconformity to occur. Therefore, we use first-order trends extracted from seismic data: a rule based on a 3D scalar field whose gradient is orthogonal to horizons; and a rule based on a seismic attribute which highlights the convergence of seismic reflectors.KeywordsSeismic DataSeismic ReflectorDynamic Time WarpingStratigraphic UnitSeismic AttributeThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.