[OpendTect_Users] [Users] Advices on Dip Steered Similarity for imaging chaotic seismic bodies

[Geert de Bruin]

Hi Nicola,

I would use the *SSIS plugin [1]*  to study the internal character of 
chaotic seismic bodies. We have seen that the chronostratigraphy and the 
chronostratigraphy-slider are great tools to study slumped/chaotic 
facies in the free F3 survey [2]. We successfully applied the *multi 2D 
workflow* [3]

Furthermore, SSIS supports a *stratal slicing* [4] workflow, which might 
give some good insights into the 3D extend of certain facies .

Another option  is the use of the neural network plugin to detect 
chaotic facies. You could create train a supervised *Neural Network* to 
create a *SlumpCube* [5]. The workflow is similar to the creation of a 
ChimneyCube [6]  but your attribute set and picks are different. The 
default attribute set might need some tweaking/extension to tailor it to 
your data.

Finally, an unsupervised *3D clustering* could be an option to isolate 
particular seismic facies [7]

I hope this will get you started.

Best regards,

Geert

[1] http://www.dgbes.com/index.php/ssis.html
[2] http://www.opendtect.org/osr/
[3] 
http://www.dgbes.com/images/stories/PDF/ssis%202d%20and%20multi-2d%20workflow.pdf
[4] ftp://ftp.dgbes.com/pub/SSIS_project/Stratal-slicing.pps
[5] 
http://opendtect.org/rel4/doc/User/dgb/chapter8.7_nn_slump_cube.htm#LINK-DEFATTR.SLUMP
[6] 
http://opendtect.org/rel4/doc/User/dgb/chapter7_applications.htm#LINK-APPLICATIONS.CHIMNEYCUBE
[7] 
http://opendtect.org/rel4/doc/User/dgb/chapter4_neural_networks.htm#LINK-NN.INTRODUCTION.UNSUPERVISED 


-- Geert de Bruin
-- Senior Geoscientist
-- dGB Earth Sciences B.V.
-- Nijverheidstraat 11-2, 7511 JM Enschede, The Netherlands
-- mailto geert.debruin at dgbes.com, http://www.dgbes.com
-- Tel: +31 53 4315155, Fax: +31 53 4315104



Nicola Scarselli wrote:
> Dear OpendTect Users,
> I was wondering if some of you have ever used deep-steered similarity
> for gaining insight into the internal features of chaotic seismic
> bodies, such as mass transport complexes and channel systems. Any
> advices how to set the parameters of the (detailed and background)
> steering cubes for having good imaging of those bodies?
> Cheers,
> Nicola
>
>