UHR3D Via Spectral Decomposition and RGB Blending
By
Brian Brookshire, Applied Science Group, NCS SubSea

 
Spectral decomposition and the subsequent red/green/blue (RGB) blending of spectrally limited sub-volumes has become a common tool in the seismic interpreter’s toolbox over the last 15 years.  Based on a quick literature search, references from as far back as 1999 are widely cited in regard to the spectral decomposition processes and applications (e.g. - Henderson, Purves, and Leppard 2007; Marfurt and Kirlin 2001; Partyka, Gridley, and Lopez 1999).  Even today, the interpretation technique is still in vogue due to advances in technologies and methodologies.  At the recent SEG International Exposition and 86th Annual Meeting in Dallas, there was an entire technical session – Spectral Decomposition Methods and Usage – devoted to this methodology (see http://library.seg.org/doi/book/10.1190/segeab.35 for a complete listing of titles and abstracts).

We had, in the past, utilized spectral decomposition techniques to create frequency limited sub-volumes for the purpose of teasing out small scale channel features in our ultrahigh-resolution 3D seismic (UHR3D) data (Brookshire 2015), but had not done much RGB blending due to the lack of a very straight-forward workflow in our chosen interpretation software.  However, we recently moved to a new interpretation software, which offers an intuitive workflow for achieving very satisfactory spectral decomposition and RGB blending results.  The following figures illustrate the steps involved in creating a stunning image of a fracture pattern found at the crest of an anticline around 1.75 seconds below seafloor (Figure 1).  To create this image, we simply picked a conformable horizon below the fractured horizon and flattened the volume to this conformable horizon (Figure 2).  We then time sliced at the fractured horizon and produced a spectrally decomposed RGB image at 50 Hz, 100 Hz and 200 Hz respectively (Figure 3).  When compared to an incoherency attribute rendering at the same time slice (Figure 4) it is obvious that, in this case, the RBG image is much sharper and more diagnostic. 


Figure 1. Profile from UHR3D seismic volume.  Fractured horizon is around the intersection of the blue and yellow lines.


Figure 2. The magenta line with purple dots represents the flattened horizon, and the thin blue line visible just above that represents the position of the time slice.


Figure 3. RGB blended image highlighting fractured horizon along the crest of the anticlinal feature. 


Figure 4. Incoherency image coincident with the RGB image in Figure 3.

References
 
Brookshire, B. N. Jr. 2015, Mass transport complex imaging with P-Cable ultrahigh-resolution 3-dimensional seismic, Near-Surface Asia Pacific Conference, Waikoloa, Hawaii, 7-10 July 2015. 113-115.
Henderson, J., S. J. Purves, and C. Leppard. 2007, Automated delineation of geological elements from 3D seismic data through analysis of multichannel, volumetric spectral decomposition data. First Break, 25, no. 3.
Marfurt, K., and R. Kirlin. 2001, Narrow-band spectral analysis and thin-bed tuning. Geophysics, 66, no. 4,1274-1283.
Partyka, G., J. Gridley, and J. Lopez. 1999, Interpretational applications of spectral decomposition in reservoir characterization. The Leading Edge, 18, no. 3,353-360.