Dear Satyendra Panwar:

Coherence is the measures of the similarity of the seismic waveform between traces neighboring or adjacent, and it reveals structural and stratigraphic changes in your data, such as faults, channels, edges of bodies in general. You can measurement the local waveform within a defined space and time aperture or window, by cross correlation of two adjacent traces, both, inline and crossline in 3D seismic data.

It is important to say that there are alternative measures of the similarity of the wavelet or waveform in seismic, such as: cross-correlation, semblance, variance, eigenstructure, gradient structural tensors.

In semblance estimate of coherence, we define the coherence such as the ratio between the energy of average traces in a defined window by the energy of the input traces. It means that you: 1- calculate the energy of input traces, 2- calculate the average wavelet within your window of analysis, 3- estimate coherent traces by their average, and 4- calculate energy of average traces.

In eigenstructure estimate of coherence, we define the coherence such as the ratio between the energy of coherent component traces in a defined window by the energy of the input traces. It means that you: 1- calculate the energy of input traces, 2- calculate the wavelet that best fits the data within your window of analysis, 3- estimate coherent compt of traces, and 4- calculate energy of coherent compt traces.

From coherence volume, you can spatially image and “directly” map faults and fracture systems without the tedious and extremely subjective method of interpreting faults on selected vertical sections, then connecting the interpreted segments to give a complete fault picture. You can more readily understand depositional systems, as the process highlights such depositional features as channels, onlap, turbidite sequences, etc.

It is very important to say to you that the quality of coherence volume attribute results depends on selection of optimum processing parameters specifying such as: dip constraints, vertical and spatial aperture, and processing algorithms.

In addition, before that, you must understand the magnitude of the acquisition seismic footprint, in focus when you working with onshore seismic, because if you do not remove or attenuate this influence, the coherence attribute highlights these changes: often they are the same order of magnitude as your target wavelet distortions related to real geological features, faults or stratigraphic. So, you need, before running the coherence attribute, try to reduce the acquisition footprint by filtering such as spectral whitening and spectral blending / FXY decon / dip enhancement. To better understand subtle fault changes, advanced structural filtering for coherency analysis that use fault preservation and across-fault smoothing techniques depending on the data requirements.

Also, because you use windows following your interesting-horizons, you must check carefully the QC of the input horizons. Sometimes you need to smooth the noisy ones.

Another consideration for your coherence attribute calculation is the fact that there are in most of the commercial software the chance to get almost instantaneously the volume. Also, that with the same input volume, you can modify the computational window options, e.g., direction of the calculus – the best perpendicular to the structural features -, the aperture window, the shape of the aperture window and size or radius, etc.

Try to read the paper attached and this landmark´s ones:

Taner, M. T., et al. 1976. Complex seismic trace analysis. Geophysics. Vol 44 No 6 (June 1979), pp 1041-1063.

Robertson, James D., and Fisher, David A. Complex seismic trace attributes. The Leading Edge, June 1988, pp 22-26.

Marfurt, et al. 1998. 3-D seismic attributes using a semblance-based coherency algorithm. Geophysics, Vol. 63 No 4 (July-August) pp 1150-1165.

Marfurt, et al. 1999. Coherency calculations in the presence of structural dip. Geophysics. Vol 64 No 1 (January-February) pp 104-111.

Best regards Satyendra,

Mario E. Sigismondi

Dear Mario,

Thank you for inside you have given to me in this topic.

I will read recommended papers and will see if those helps. Further to this topic i have a query regarding dip calculation. As per my knowledge ,dip is required to calculate coherency in Eigenbased method and in semblance based dip scanning method and in the algorithm dip is simply taken for maximum coherency in that particular window which is may not be correct dip. Say our area/data consist noise which is as good as any event and difficult to remove by any mean of processing. These noise can be of nature of conflicting dips or anything. so my query is , In those scenarios how successful these methods would be ?

Thank you in advance!

regards,

Satyendra Panar,

I can only praise you for wanting to do this yourself. I have a suggestion, why not look at the programs in "open source" seismic interpretation packages? Wikipedia lists a long comparison of free geophysical software. It may be a challenge to find the right programs free and open, but then you would have something you could adapt.

Regards

Jim Ballard

Thank you Jim for suggestion. I am looking for open source codes.

regards,

Satyendra

Article 3-D seismic attributes using a semblance-based coherency algorithm

http://www.searchanddiscovery.com/pdfz/documents/2018/42273chopra/ndx_chopra.pdf.html

Dear Satyendra Panwar ,

I bumped into your question while I was searching for something similar. I don't know if you figured it out by now, but this tutorial

https://github.com/seg/tutorials-2015/blob/master/1512_Semblance_coherence_and_discontinuity/Discontinuity_tutorial.ipynb

seems interesting and related to your question.

Regards