Publications

Spectral decomposition is an established technique used for layer stratigraphic feature visualization (Marfurt and Kirlin, 2001), thickness determination (Partyka et al, 1999) and as direct hydrocarbon indicator (Castagna et al., 2003, Sinha et al., 2005). The analysis often revolves around defining an explicit window using interpreted surfaces. This introduces interpreter bias as optimal window, and by extension, the viability of interpreted surfaces plays an intense role in final output and interpretation. This work attempts to show time frequency analysis can be made unbiased and optimized using frequency gathers. Continuous Wavelet Transform (CWT) technique has been implemented for spectral decomposition analysis independent of interpreted surface. A 3D seismic time migrated volume is used as input to generate spectrally decomposed signal envelope frequency gathers for identifying frequency bands sensitive to feature under study. The sensitive frequency band information is then utilized to generate spectrally decomposed signal envelope volumes for 3D visualization and analysis. Comparative study of sensitive and non-sensitive frequency bands carried out in this work illustrates the effectiveness of frequency gathers. The time frequency gather generated on the studied feature shows sensitivity of the feature to certain frequency bands and non-sensitivity to other bands. This paper describes frequency gathers as suitable alternative to spectral decomposition analysis based on interpreted surfaces, particularly when interpretation can be subjective. It helps in optimizing the number of frequency volumes that need to be generated for effective and unbiased analysis of any feature whether for stratigraphic visualization or thickness variation. This paper proposes and demonstrates a new approach for performing robust spectral decomposition analysis, which enables user to focus on relevant frequency bands and remove non-consequent bands from analysis, saving time and resources. This results a more detailed understanding of reservoir dispersal pattern and geomorphic features for un-biased understanding of reservoir geology.