[OpendTect_Users] Sweetness Attribute

Mon Jan 27 17:48:52 CET 2014

Hi Rene and others,

I believe my colleague (Friso Brouwer) has already furnished the concept &
implications very well.

Some other ideas that came across are below:

1- Try cross-plotting the two attributes along a given event that you are
familiar with. See what comes up. This is a statistical way to look at the
things. I am certain that you will find minor statistical differences due
to the fact that has been earlier explained by Ross.

2- Try *Spectral Decomposition *on the same horizon/volume. The advantage
of this approach would be to recognize the freq. response generated by a
specific feature. You may also cross-plot the sweetness (x-axis) vs spectra
decomp (y-axis) to see the trend and clusters. Note that in OpendTect you
can link the cross-plot with the scene by displaying the extracted point
data in the scene. This will help you to locate the odd responses (such
tuning effects, anomalies of any kind, geologic feature or noise).

3- HorizonCube density is a stratigraphic attribute that has recently been
introduced by us. You may want to use such an attribute to describe the
geometrical meaning behind the sweetness. Here you can certainly create
geobodies of good and bad sweetness by combining various attributes
(sweetness, spectral decomp, HC density, ...). You may use if/then
statements or Neural Networks.

4- Try any other seismic attribute that is not dependent on frequency or
amplitude. This is hard though as these two are the primary components of a
reflected wave/sound.

*5- *Plz don't mind if I close my e-mail like this,

*''Don't believe any thing & keep on evaluating until death! Cheers :-)''*

Best regards,

*Farrukh*

On Mon, Jan 27, 2014 at 2:12 PM, Friso Brouwer <friso.brouwer at dgbes.com>wrote:

> Jack:
>
> I do not know of an online resource, but basically the sweetness attribute
> would combine all pitfalls found in amplitude and frequency analysis (note
> that sounds worse than it is, if you know what you are doing, sweetness is
> a really useful attribute).
>
> Typical pitfalls are:
> Amplitude term: being in a dim spot regime (sweetness will not work and
> you are missing reservoir), false positives due to hard streaks
> (carbonates, cemented sandstone, evaporates,..), high amplitudes due to
> volcanic derived sediments, etc etc. Some you can screen for some of these
> problems, for example hard streaks are supposed to have an opposite
> polarity. Hilterman's DISC is probably excellent reading on this subject.
>
> Frequency term: bad processing, gas chimneys (rather than gas filled
> reservoir), reflections of rugose reflection surfaces (salt, erosional
> surfaces), converted waves not accounted for. Again more found in
>  literature.
>
> So if you find an sweetness feature of interest, don't just assume it is
> good, investigate further: know the regional stratigraphy, look at the
> polarity, try to reproduce using forward modelling, create a HorizonCube
> slice through the feature and investigate it's morphology, check its
> downdip  limit (and consistency thereof) if you suspect sweetness to be
> hydrocarbon related, etc etc. Note this can be done very easily in
> OpendTect.
>
> Hope this helps.
>
>
> On Sat, Jan 25, 2014 at 12:20 AM, Jack Barajas <jack.barajas at gmail.com>wrote:
>
>> Is there an online resource discussing the pitfalls if sweetness 1 and 2?
>> —
>> Jack
>>
>>
>> On Sat, Jan 25, 2014 at 12:57 PM, friso.brouwer <friso.brouwer at dgbes.com>wrote:
>>
>>> Rene,
>>>
>>> The sweetness attribute is a plain and simple tool to highlight 1) thick
>>> clean reservoir, and/or 2) hydrocarbons (HC) filled reservoir.
>>>
>>> Principle is you divide a amplitude attribute with a frequency attribute.
>>>
>>> For reservoir (low AI sand embedded in shale) the effects are that when
>>> the reservoir gets cleaner or thicker, the amplitude term increases due to
>>> higher reflectivity and frequency of the combined reflections lower as the
>>> thicknesses increases.
>>>
>>> For HC presence (assuming an AVO 3 environment, where introduction of HC
>>> leads to brightening), the effects are when HC gets introduced that
>>> amplitude increases due to higher reflectivity and frequency of the
>>> reflection is lower due to the combined effects of HC on the frequency
>>> content.
>>>
>>> From the mathematical formulation you see that all these effects work in
>>> the same direction. Of course it is also a very simple attribute and there
>>> are lots of pit-falls, which I will not list here, as it is a bit besides
>>> your question.
>>>
>>> The inst. amp / sqrt(inst. freq) is the original definition of this
>>> attribute, but it is easy to understand you can stick different terms in
>>> the equation for the amplitude related numerator and frequency related
>>> denominator and stay true to the underlying concept. The results will be
>>> strongly related, but not equal. I often use the instantaneous amplitude in
>>> the numerator as it removes the question what the best time gate, but I do
>>> not like to use the instantaneous frequency as it is sensitive to thin beds
>>> (which is, by the way, the basis of Tury Tanner's thin bed attribute).
>>> Putting a square root operator in the denominator compresses its original
>>> dynamic range and limits the effect of the higher frequencies.
>>>
>>> The nice thing of the attribute is that you can do other things as well,
>>> which are all quite easily implemented using OpendTect's mathematics
>>> attribute and other attributes:
>>> * Using energy instead of rms or inst. ampl in the numerator, which
>>> expands the dynamic range of the amplitudes and focuses the results on the
>>> very high amplitudes only.
>>> * Replace the amplitude term with a far stack amplitude or an AVO
>>> attribute.
>>> * Shift the frequency range, such that, with the original range going
>>> from Fmin to Fmax, the shifted range goes from c to Fmax-Fmin+c (with 0< c
>>> << Fmin), this will really boost the effect of low frequencies in the
>>> sweetness attribute response.
>>> * Shift the time gate of the frequency attribute, such that the
>>> frequencies are extracted below the evaluation point. This is very useful
>>> as most frequency effects due to HC are found below the reservoir
>>> reflection.
>>>
>>> So it is a really nice attribute to play with and optimize doing
>>> recognizance type evaluation of reservoir and HC presence, but be cautious
>>> in its interpretation, as there is a long list if pitfalls.
>>>
>>>  I hope this helps you forward.
>>>
>>>
>>>
>>>>
>>>> ---------- Forwarded message ----------
>>>> From: René Grobe <grobe at geo-t.de>
>>>> Date: Fri, Jan 24, 2014 at 2:17 PM
>>>> Subject: [OpendTect_Users] Sweetness Attribute
>>>> To: users at opendtect.org
>>>>
>>>>
>>>>  Hi All,
>>>>
>>>>
>>>>
>>>> as non-geophysicist I have a question about the definition of the
>>>> “Sweetness” attribute. In literature the sweetness attribute seems to be
>>>> usually mathematically defined as “Sweetness(1) = Instantaneous
>>>> Amplitude/Sqrt(Instantaneous Frequency)”. In the OpendTect Attributes
>>>> Matrix (https://opendtect.org/opendtect_attributes_matrix/)
>>>> “Sweetness(2) = RMS Energy/Average Frequency” or “RMS Amplitude/Average
>>>> Frequency”. It is clear that both definitions - Sweetness(1) and
>>>> Sweetness(2) - generally relate Amplitude to Frequency, but do they exactly
>>>> produce the same result or do they differ in any way? If so, how can one
>>>> describe the difference and its impact on interpreting the result?
>>>>
>>>>
>>>>
>>>> Thank you very much!
>>>>
>>>> Best regards,
>>>>
>>>> René
>>>>
>>>>
>>>>
>>>> _____________________________________________
>>>>
>>>>
>>>>
>>>> René Grobe, PhD.
>>>>
>>>> Geologist
>>>>
>>>> GeoThermal Engineering GmbH (GeoT)
>>>>
>>>> Baischstr. 8
>>>>
>>>> D - 76133 Karlsruhe
>>>>
>>>> Tel:         +49 721 570 44 6 811
>>>>
>>>> Fax:        +49 721 570 44 6 89
>>>>
>>>> Mail:      Grobe at Geo-T.de <grobe at geo-t.de>
>>>>
>>>> Web:     *http://www.Geo-T.de/ <http://www.geo-t.de/>*
>>>>
>>>> _____________________________________________
>>>>
>>>>
>>>>
>>>> Geschäftsführer: Dr.-Ing. Horst Kreuter, Registergericht Mannheim HRB
>>>> 110969
>>>>
>>>>
>>>>
>>>>
>>>> *Diese E-Mail enthält vertrauliche und/oder rechtliche geschützte
>>>> Informationen. Wenn Sie nicht der richtige Adressat sind oder diese E-Mail
>>>> irrtümlich erhalten haben, informieren Sie bitte den Absender und löschen
>>>> Sie diese Mail. Das unerlaubte Kopieren sowie die unbefugte Weitergabe
>>>> dieser E-Mail und der darin enthaltenen Informationen sind nicht gestattet.
>>>> **This e-mail may contain confidential and/or privileged information.
>>>> If you are not the intended recipient (or have received this e-mail in
>>>> error) please notify the sender immediately and delete this e-mail. Any
>>>> unauthorized copying, disclosure or distribution of the material in this
>>>> e-mail is strictly forbidden.*
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> Users mailing list
>>>> Users at opendtect.org
>>>> http://lists.opendtect.org/mailman/listinfo/users
>>>>
>>>> You receive this mail because you are listed on users at opendtect.org To
>>>> unsubscribe please go to
>>>> http://lists.opendtect.org/mailman/options/users or click here :
>>>> http://lists.opendtect.org/mailman/options/users/paul.degroot%40dgbes.com. If you encounter any problems, please contact
>>>> support at opendtect.org .
>>>>
>>>>
>>>>
>>
>
> _______________________________________________
> Users mailing list
> Users at opendtect.org
> http://lists.opendtect.org/mailman/listinfo/users
>
> You receive this mail because you are listed on users at opendtect.org To
> unsubscribe please go to http://lists.opendtect.org/mailman/options/usersor click here :
> http://lists.opendtect.org/mailman/options/users/farrukh.qayyum%40dgbes.com. If you encounter any problems, please contact
> support at opendtect.org .
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opendtect.org/pipermail/users/attachments/20140127/ee82e8e2/attachment-0001.html>