Demo of using the Geo5 FracTex plugin in OpendTect to compute texture attributes and analyze fracture anisotropy and orientations from seismic data. This demo shows how to extract seismic texture, apply directional attributes, and evaluate fracture dip azimuths for fault and reservoir characterization.

#OpendTect #Geo5 #Fractex #SeismicAttributes #Geophysics #FaultAnalysis #ReservoirCharacterization #SeismicInterpretation

Duration: 7:52

--- Transcript ---

And FracTex is in this over here so also here we have the same dataset as you can see the same Horizon the same Thinned Fault Likelihood overlay in this case beefed up a little bit by Computing the RMS of tfl. And we have a small seismic section I first go into my attribute window in which I have defined a number of attributes if I first look at the texture attributes you can see here we have groups of attributes if I look at all the attributes. Then go to texture attributes we have two inputs we have texture and texture directional attributes so these are the basic ones.

And these are the ones that have been added by Geo5 and in which you can compute texture in different directions let's first look at the basic one texture you have to specify an input seismic you specify how much seismic you want to analyze around this position in this case the time gate is minus 8 milliseconds. Above 8 milliseconds below you step out three samples in the inline Direction on both sides so that is a seven samples input same in a crossline direction. And you can do steering or not steering if you do steering then you will follow the seismic reflectors better than if you don't do the steering you're just.

Then extracting the information horizontally here you specify which glcm attribute you want to compute and you specify whether you want to compute a matrix based on 16x16 gray levels or 32x32 you need to compute the input data range the minimum. And the maximum because that will then be transformed into gray levels and then you get in this case a name you assign it. And it becomes an attribute that you can use and play with so I already played with this attribute.

And I showing it here this is the texture contrast and that looks like this and if we. Now look at directional attributes and it's slightly more or inputs in the and outputs in the parameters here. So we again compute the number of gray levels we need an min max of the input data which can be computed automatically we Define the window in lines.

And cross line steps and vertically and then here you can see we have many more attributes that can be computed in this particular case unique of course to this particular analysis is that you can specify the direction in which you want to compute these different attributes. And whether you want to use steering or not you give it a name and you add it. And you compute it and I've done that here and it looks in this particular case like this for this particular attribute the directional correlation in all directions it may not look that interesting.

But it becomes interesting of course if you are going to analyze it in FracTex because in FracTex we have for instance this attribute which is FracTex energy. And isotropy you have to open it to get more parameters here you have the number of directions in which you can compute. And you can increase that but then you also have to increase the window analysis size the computation times become pretty large if you do that.

So typically you would use 13 directions we have these attributes that we can select now dlcm attributes. And then we can output two attributes one is the anisotropy factor and the second one is the fracture dip azimuth. So the anisotropy factor again is the value from one or higher and the higher it is the more an isotropic the image is at this location.

And the fracture dip azimuth is the estimate of the largest variation in an anisotropy and that should correspond to the fracture orientation again you can do this full steering you give it a name. And then here I have the different attributes fracture contrast energy inverse distance moment entropy and it's all an isotropy that I'm outputting. And here I'm outputting the corresponding fraction dip estimates for the same attributes now this calculation takes quite some time.

So I have computed a few up front and this one for instance is the fracture FracTex homogeneity anisotropy. And we can see that the highest anisotropy is in this area if we compare this with the RMS of the tfl. Then we can see yes not surprisingly that there's a lot of an isotropy here.

And actually this is an artifact it's a typical Edge effect we see with tfl calculations it's not just our software this all softwares have that. And we actually have ways to minimize this so but it's still there and that is. Also reflected here in the FracTex and isotropy calculation but over here where we have the real faults it.

Also scores high in terms of an isotropy so this would be interesting to further evaluate and more interesting I think is the fracture dip azimuth which looks like this. And this picks up indeed default directions quite nicely so this is one of the major outputs main outputs of the FracTex analysis. And this you can do for all these different attributes and if all goes well then yeah you pick up the an isotopy in the image in this case it is related to faults because there are real faults in the data.

But if you are looking at a fractured Reservoir then there might be not any displacement but still there might be anisotropy due to the fractures in the system. And that is the end of my presentation so thank you very much.