Qualitative and quantitative prediction based on seismic data (by the example of Tyumen formation in Western Siberia)

The authors discuss the problems related to qualitative and quantitative prediction of reservoir properties of the Tyumen horizons. The paper covers the main challenges of the Tyumen formations studies using seismic data, which result from poor distinction between their lithotypes in the elastic parameter field. Particular attention is paid to the specifics of identification of geological objects having paleo-riverbed genesis using the results of frequency decomposition with RGB blending methodology. They made the basis for an attempt to classify paleochannels into following two types on the basis of several indicators: filled mainly with sand and with clay material. In order to verify the assumptions made, forward modelling was carried out (with the use of actual well data) for paleochannels with different types of filling. The obtained results of forward modelling together with well data (well logging, core) allowed formulating criteria for classification of identified paleochannel objects into promising (filled with sand) and showing no promise (shaled-out). The authors demonstrate the results of paleochannel body prediction by the examples from one of the fields in Khanty-Mansiysk Autonomous District, which made a basis for building a conceptual geological model and a sedimentation scheme of Yu3 reservoir