APPLICATION OF WAVEFORM CLASSIFICATION AND MULTI-ATTRIBUTE ANALYSIS TO MAP DIFFERENT LITHOFACIES: CASE STUDY FROM VULCAN SUB-BASIN, AUSTRALIA

Abstract

Modern seismic techniques such as waveform classification and multi-attribute analysis can define facies and reservoir parameters with detail than traditional time and amplitude mapping. This report tries to establish link between lithology and seismic waveform classification by using 3D seismic and well data of Vulcan Sub-Basin, North West Shelf, Australia. Moreover, multi-attribute analysis was also performed and used in combination with waveform classification to predict different lithofacies in the area. At first step, maps were produced through unconstrained or unsupervised classification using different number of classes varying from five to ten. These maps are seismic data driven and without any guidance from well data. In the second step, constrained or supervised classification uses the known information at well locations. Facies maps of three reservoir intervals of Montara, Plover and Nome formations were computed based on both supervised and unsupervised classification. Waveform classification successfully mapped facies such as limestone, cemented sandstone and sandstone. In order to improve the prediction, multi-attribute analysis was applied to predict GR and density for key reservoir intervals. Horizon slices of GR and density predicted volumes differentiate limestone and sandstone. Combining waveform classification and multiattribute analysis detail depositional environment for three reservoir formations can be predicted. Pre-rift Triassic -Nome Formation indicates shallow marine and shelf margin environment. Plover Formation is characterized by fluvial sedimentation of sand and shale. This formation is eroded in some parts due to tectonic uplift. Uppermost studied reservoir of Montara is interpreted as regressive to transgressive shoreface-delta.