Facies trend modeling for SAGD application at Surmont

Large scale trends are important geological features that should be accounted for in geostatistical characterization of uncertainty. The workflow for uncertainty characterization accounting for a trend involves two major steps: (1) trend modeling and (2) integrating the trend into geostatistical estimation and/or simulation. While there are several options to consider for integrating trend models into geostatistical workflow, building a reasonable and fair trend model based on data coming from different sources and/or defined on different support still presents a challenge. In this paper we describe a locally varying methodology for building a facies proportion trend for Surmont area of Athabasca Oil Sands. The trend is built based on vertical well data and seismic attribute – acoustic impedance; several probability combination schemes are considered for merging facies probability cubes calculated from the two data sources. All steps required building trend models and selecting one representative based on the criteria of fairness and certainty are documented and detailed.