A new method for Spectral Decomposition is featured in an article by its inventor, Geomodeling’s Senior Geophysicist Rongfeng Zhang, in the Geophysical Corner of AAPG Explorer.  Rongfeng’s article, “Effective Ways to Eliminate Side-Lobe Effects” describes some of the issues associated with Spectral Decomposition that can result in self-related events on images and a new method for addressing this phenomenon

Dr. Zhang’s method for Spectral Decomposition, Real Seismic Wavelet Transform (RSWT), uses a real wavelet extracted from the seismic data instead of a ideal wavelet derived from a mathematical expression (the technique used in most Spectral Decomposition methods).  The extracted wavelet is numerically squeezed and stretched, and then convolved with seismic trace. The result shows much less ringing or side lobe effect than the result from Spectral Decomposition using a Continuous Wavelet Transform (CWT) but has the same high resolution characteristics as CWT.

The new RSWT method removes side-lobe effects and, as shown in the pictures above and below, can produce improved definition in areas of channel and valley fill.

Comparison of Spectral Decomposition Methods

Rongfeng has presented talks on his new method at both Geoconvention 2013 in Calgary and SEG 2013 in Houston:

Geoconvention 2013:  Spectral Decomposition with Real Seismic Wavelet
(extended abstract available directly from this link)

SEG 2013:  Spectral Decomposition with Extracted Seismic Wavelet
(extended abstract available through SEG from this link)

This new RSWT method for Spectral Decomposition will be included in AttributeStudio 7.4, to be released later this month.  If you would like to see it in action, come by our booth (#1857) at AAPG in Houston, April 6-9, 2014 or contact us using the form below to schedule a presentation and demonstration.

 – John Sherman, Vice President of Marketing and Product Development