Abstract

Traditional seismic interpretation methods that rely on time mapping and amplitude analysis often struggle to distinguish subtle variations in geological facies. Analysing seismic waveform shape and character provides a more sensitive method for identifying these differences. This paper focusses on seismic waveform classification into different classes using neural network and the classified waveforms are then correlated with known geological facies. Waveform classification can also be combined with multi-attribute analysis to improve facies discrimination. Multi-attribute seismic facies maps offer better insight into lithology distribution and depositional environments, aiding in the identification of key reservoir rocks. This classification provides a qualitative method for capturing meaningful variations in facies. This integrated method shows improved performance over traditional techniques in detecting subtle facies differences, providing valuable information for hydrocarbon exploration