[Users] How-to: Display and store thickness maps

[Arnaud Huck]

Dear Sir,
Indeed, what you decribed is often requested, i.e. computing attributes 
between two surfaces rather than in a fixed time-window.
You only need to modify two steps of the previous workflow:

1) The first attribute "Blank outside" needs to be computed and stored 
between two horizons. For this use the "Processing menu" in the title 
bar of OpendTect. Select the option "Create output using horizon", and 
then the sub-item "Between horizons". The quantity to output will be the 
"Blank outside attribute", you have to specify the top and base of your 
interval by two horizons, which can be extended with fixed time shifts. 
The "Value outside computed area needs to be set to 1e30, the undefined 
value, and the last step is to provide a name for you new volume.

2) Lets say that the difference between your two horizons is going from 
A (minimum) to B (maximum), in time or depth. If you want to see the 
result on you top horizon, you need to set the time gate of the second 
attribute to [0, B]. If you want to see the result on your base horizon, 
the time-gate needs to be set to [-B, 0]. Then, even if you go outside 
of the region delimited by your two horizons, the corresponding samples 
are set to undefined and thus will not contribute to your sum. The key 
step here is to re-direct the input of this second attribute from the 
"Blank oustide" attribute (previous workflow), to your stored volume 
created in the updated first step (new workflow).

3) The third attribute does not change, as it is just a conversion.

Remark: The procedure described in the first step in very convenient to 
mute the regions outside your interval of interest. It works for any 
stored volumes, attributes and neural network output.

Best regards,
Arnaud Huck.

-- dGB Earth Sciences B.V.
-- Nijverheidstraat 11-2, 7511 JM Enschede, The Netherlands
-- mailto: arnaud.huck at dgb-group.com, http://www.dgb-group.com
-- Tel: +31 53 4315155 , Fax: +31 53 4315104



Jianyong Hou wrote:

> Dear Sir
> You told us a useful method to calculate the thickness of a 
> interesting layer. But I have a question: In your second step, a time 
> gate is used to delimit the interesting layer. However, when the layer 
> is inclined not gentle, it is difficult to extract only a layer in a 
> regular time gate. Some layers locating upper and/or below the target 
> layer and with similar lithology, or simialr reflective features will 
> be involved in the time gate. If so, the summed thickness can include 
> thickness of the upper and/or below layers. So, my question is: in 
> OpendTect, if can you delimit the region with two horizons upper and 
> below the target layer not to delimit it by a regular time gate?
>
> Best regards,
> JGI Jianyong Hou
>
>
>
> ----- Original Message ----- From: "Arnaud Huck" 
> <arnaud.huck at dgb-group.com>
> To: "dGB OpendTect userlist" <users at opendtect.org>
> Sent: Wednesday, January 24, 2007 10:15 PM
> Subject: [Users] How-to: Display and store thickness maps
>
>
>> Dear OpendTect user,
>> A common way at using inverted volumes is to extract thickness maps. 
>> From an inverted impedance volume, and based on well analyses to 
>> define cut-offs, it is possible to separate sands from the background 
>> impedance data.
>>
>> Let's assume such an analysis would have told us that (absolute) 
>> impedances lower than 9.000 (m/s)x(g/cc) are related to sands. In 
>> OpendTect, you can derive from such a cut-off thickness maps in less 
>> than five minutes with the help of three combined attributes:
>>
>> 1) At first, you need to enter you cut-off in a attribute of type 
>> "Mathematics". The formula would be the following (if-then-else 
>> statement):
>> x0 < 9000 ? 1 : 1e30
>> where x0 is your impedance volume.
>> This attribute is acting like a flag, giving us the sample locations 
>> where impedance is lower than our threshold, while the other samples 
>> are set to 1e30, which is our undefined value.
>> We can call this attribute "Blank outside".
>>
>> 2) Secondly, the thickness map itself is drawn by summing the number 
>> of samples set to 1 in our region of interest. This region is 
>> delimited by a time-gate, relative to the horizon on which we will 
>> apply this attribute.
>> For this purpose, we use an attribute of type "Volume statistics". 
>> The input data is the previous attribute "Blank outside", you can 
>> enter your desired time-gate. The stepout is 0 / 0 (trace-by-trace 
>> process), Output statistic is "Sum", and Steering is set to "None" as 
>> we will apply it on an horizon. Undefined values are not contributing 
>> to the sum.
>> We can call this attribute "Sum thickness", and returns a number of 
>> (flagged) samples
>>
>> 3) The third attribute aims at converting the result of the second 
>> attribute in meters or seconds.
>> If your impedance volume is in depth, you just need to multiply the 
>> previous attribute by the depth sampling rate.
>> In a more normal situation, your impedance cube is in time, and then 
>> we need to multiply the "Sum thickness" attribute by the (time) 
>> sampling rate, 4 msec. for instance, and by the average interval 
>> velocity within your formation. This is done with a second 
>> "Mathematics" attribute, with the kind of formula:
>> x0*0.004*2000.0
>> where x0 is the "Sum thickness attribute", 0.004 is the sampling rate 
>> in seconds, and 2000.0 is the interval velocity in m/s.
>> The thickness returned would then be in meter. We can call this 
>> attribute "Thickness in m." for instance.
>> Optionnaly, you can also read the velocity from another volume if 
>> available.
>>
>> Finally, you can get your thickness map by applying the "Thickness in 
>> m." attribute allong your horizon. Please note that you can also use 
>> this approach with lithologies/facies cubes.
>>
>> If you have any questions, do not hesitate to ask me.
>>
>> Best regards,
>> Arnaud Huck
>> Geoscientist.
>>
>> -- dGB Earth Sciences B.V.
>> -- Nijverheidstraat 11-2, 7511 JM Enschede, The Netherlands
>> -- mailto: arnaud.huck at dgb-group.com, http://www.dgb-group.com
>> -- Tel: +31 53 4315155 , Fax: +31 53 4315104
>>
>>
>> _______________________________________________
>> Users mailing list
>> Users at opendtect.org
>> http://lists.opendtect.org/mailman/listinfo/users
>>
>>
>
>
>