[OpendTect_Users] Scaling multiple angle-stack volumes

[Christopher Ross]

Arnaud,

While this approach will normalize the amplitudes between the two stacks (in this case), it does not preserve the AVO nature of the responses which will vary temporaly and spatially. To address this issue you need to examine the gathers and determine the decay rate of the background reflectors, and address the issues before stack. See Ross & Beale in Geophysics (Circa 1994) for a better example. Normalizing the near and far offset stack to be the same assumes that all of the background reflectors have a essentially a zero-gradient. Perhaps they do, but that needs to be determined via well control or some other method before hand.

Christopher P. Ross
Geophysicist
830.896.7656
-----Original Message-----
From: Arnaud Huck [mailto:arnaud.huck at dgbes.com]
Sent: Monday, November 21, 2011 02:35 AM
To: 'Tom Kearney'
Cc: users at opendtect.org
Subject: Re: [OpendTect_Users] Scaling multiple angle-stack volumes

 Dear Tom,

 The optimal way to assess the amplitude level of a volume is to compare RMS amplitude maps over a specific interval. This task can be done with the "Stratal Amplitude" module, to be found in the menu "Processing --> Create Grid output --> Stratal Amplitude". You can select here a stored volume or attribute, and compute its statistics in an interval defined by one or two horizons and a time gate. For instance you would here output the RMS.

 You need to perform the operation for both stacks, that will provide two grids stored along the same horizon. To assess the difference you need to make the grids available as attributes. That is the purpose of the "Horizon" attribute, so you need to define two attributes of type Horizon, that will each output one of the two grids. The input volume specifies the 3D area where the attribute can be applied (its amplitudes are not used).
 Finally you can compare both RMS grids by computing the difference or the ratio between the two horizon-grid attributes. All three attribute can be applied on any horizon, or alternatively on time-slices. Use the histogram tools of the attribute layer to get survey wide statistics, like the mean value of all RMSfar/RMSnear.
 The scaling is to be applied with the mathematics attribute or in the Copy module of the volume manager.


 The extraction of the amplitude spectrum in an horizon-defined gate requires similarly a few steps:
 1. First mute the data outside the target interval: Create a new volume in Processing --> Create Seismic Output --> Between horizons. Select the input volume, target interval and set the "Value outside computed area" to undef instead of zero (note that this is only important if the mean of your input is not equal or close to 0).
 2. If needed: Use to volume manager to mute the seismic laterally, outside a given outline (polygon).
 3. Then add a volume element to the tree, extend it such that it covers the lateral and vertical extend of the data.
 4. Right-click on the volume element and use "Show amplitude spectrum".
 Note that depending on the size of the dataset this workflow can take quite some time because you are loading all the traces in memory. It would be much faster for a very similar result to pop-up the amplitude spectrum of many, single vertical slices. Finally you may also want to try to use to extract statistical wavelets and compare their amplitude spectra. The wavelet extraction tool will do all four steps in one go and you can control the lateral decimation level by increasing the inline/crossline stepout for the calculation.

 Best regards,
 Arnaud Huck.
-- Senior Geoscientist-- dGB Earth Sciences-- Nijverheidstraat 11-2, 7511 JM Enschede, The Netherlands-- mailto: arnaud.huck at dgbes.com, http://www.dgbes.com-- Tel: +31 53 4315155 , Fax: +31 53 4315104
 On 11/21/2011 04:46 AM, Tom Kearney wrote: All,

I am trying to use Opendtect to analyze angle stack volumes for AVO effects. However, before I start, I am conscious of the need to match the frequency and amplitude spectra of the volumes. I notice a consistent amplitude difference between my near-angle volume and far-angle volume, which is nearly always of lower amplitude, irrespective of any AVO effect. Is there a workflow whereby I can scale the volumes to be consistent? Ideally I would like to create an amplitude spectrum of all the traces in a particular inline-xline-TWT range, but cannot find a way to do this.

Thanks,

Tom Kearney.



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opendtect.org/pipermail/users/attachments/20111121/3de76d70/attachment.html>