Dr. Tom Smith presenting on Machine Learning at the 3D Seismic Symposium on March 6th in Denver
What is the "holy grail" of Machine Learning in seismic interpretation? by Dr. Tom Smith, GSH Luncheon 2018
Using Attributes to Interpret the Environment of Deposition - A Video Course. Taught by Kurt Marfurt, Rocky Roden, and ChingWen Chen
Dr. Kurt Marfurt and Dr. Tom Smith featured in the July edition of AOGR on Machine Learning and Multi-Attribute Analysis
Rocky Roden and Ching Wen Chen in May edition of First Break - Interpretation of DHI Characteristics using Machine Learning

Smoothed Frequency

Smoothed Frequency is the time-average of the Instantaneous frequency. Since instantaneous frequencies are influenced by the bed thickness, it is best to observe them without too much interference. One approach to addressing the reversals is accomplished by applying a time-average smoother to the frequency which will smooth through the reversals.

The main use for this attribute is in calculation of the Thin Bed attribute which is the difference between the instantaneous and time-weighted frequencies.

Here is a display of frequency in blue, smoothed frequency in green and thin bed tb(t) in red. All three are included here as the thin bed is the difference between the frequency and the smoothed frequency and it makes sense to show all three.
SFreq - 03.png

Interpretation Use: 

  • Corresponds to the average frequency (centroid) of the amplitude spectrum of the seismic wavelet
  • Seismic character correlator
  • Indicates the edges of low impedance thin beds.
  • Hydrocarbon indicator by low frequency anomaly. The unconsolidated sands, due to the oil content of the pores, sometimes accentuate this effect.
  • Fracture zone indicator appear as low frequency zones
  • Chaotic reflection zone indicator
  • Bed thickness indicator. Higher frequencies indicate sharp interfaces or thin shale bedding; lower frequencies indicate sand rich bedding
  • Sand/shale ratio indicator

Recommended Colorbar:   

This attribute usually has a Gaussian distribution centering around a dominant frequency.  Using a lighter color for the dominant frequency will allow highlighting of lower and higher frequencies if desired.  Often bands of colors can be used to highlight specific ranges.  Here we are using the same colorbar as we did for the instantaneous frequency

 Example colorbar and amplitude spectrum

Example colorbar and amplitude spectrum

Example:

 Vertical display of Smoothed Frequency

Vertical display of Smoothed Frequency

 Time Slice of Smoothed Frequency

Time Slice of Smoothed Frequency

 Frequency Spectrum

Frequency Spectrum

Computation: 

M. T. Taner introduced an envelope weighted instantaneous frequency which is smoothed over time window Ͳ. His smoothed instantaneous frequency is given as
SFreq - 08.PNG

References:

Comment