A set of seismic attributes has been computed from 3D time migrated seismic data in Waitara prospect off New Zealand covering an area of ~122 sq.km and amalgamated them with the interpreter’s acquaintances based on artificial neural network (ANN). This has resulted into a meta-attribute, defined as the thinned fault cube (TFC), that augments interpretation of geological discontinuities from seismic data. First of all, we condition the data to make geologic features free from noises, then extract suitable attributes, and train them over example locations selected from the data volume through a fully connected multi-layer perceptron (MLP). It is observed that the TFC efficiently illuminates the geological discontinuities and sharpens the fault images
(a) The TFC meta-attribute, co-rendered with amplitude data, along the seismic inline 1078 showing distinct subsurface faults (left panel). (b) The TFC meta-attribute at a time horizon (t = 0.88 s), marked by yellow line in left panel, co-rendered with the amplitude data, has captured structural discontinuities and reveals enhanced geological features for interpretation.
For more details, please visit
P. Chinmoy Kumar & K. Sain, 2018. Attribute amalgamation-aiding interpretation of faults from seismic data: An example from Waitara 3D prospect, offshore Taranaki basin, New Zealand, Jour. of App. Geophys., 159, 52-68.