[SOLVED] dirvect in vario.pgs

Here you can report the problems encountered when using a function relative to the variography:
- experimental variography (vario.calc, vario.grid, vmap.calc, ...)
- model fitting (model.auto)

[SOLVED] dirvect in vario.pgs

Postby twebberbr » Fri Jul 29, 2016 5:19 pm

Bonjour folks.

Hope everything is fine with you.

I 'm trying to use the vario.psg tool but I have some questions about the "dirvect".

I have a 3D dataset and I want to perfom the variogram calculation in three direction. The major have an azimut of 155º, second is 65° and the minor is normal to the horizontal plane. I just don't understand how to inform this using the "dirvect".

My lithotype rule is very simple, with only two lithotypes.
rule.PNG
rule.PNG (7.8 KiB) Viewed 5932 times


And the data is as follows.
dados.PNG
dados.PNG (11.39 KiB) Viewed 5932 times


The field 5 is the lithotype, and the proportions are the fields 6 and 7.

It's correct?

Thanks for the attention.

Tiago
Last edited by twebberbr on Wed Aug 17, 2016 1:52 pm, edited 1 time in total.
twebberbr
 
Posts: 1
Joined: Tue May 10, 2016 2:09 pm

Re: dirvect in vario.pgs

Postby Didier Renard » Tue Aug 02, 2016 8:02 pm

Hello Tiago,

Good to hear from you.
I understand that your data set is defined in 3-D. You want to calculate the variogram (of the underlying GRF, using vario,pgs)
in given directions.
The definition of these directions would be the same in the case of calculating the experimental variograms
in several directions in the 3-D space ... using vario.calc() procedure.

In all cases, we must define the 'dirvect' argument.

As mentioned in the documentation (common to thee two functions), dirvect is a matrix which defines the set of directions in which the variogram must be calculated. Its number of rows correspond to the space dimension (i.e. 3) and the number of columns gives the number of directions.

As mentioned, this matrix can be generated by using the function get.directions() which is more flexible to generate this matrix.
In particular, it contains a possibility for defining a set of directions in the XoY plane for a 3-D space:

Code: Select all
dirvect = get.directions(dirvect=c(155,65),ndim=3)


This provides the following result:
Code: Select all
          [,1]      [,2]
[1,] -0.9063078 0.4226183
[2,]  0.4226183 0.9063078
[3,]  0.0000000 0.0000000


which obviously corresponds to 2 directions (because of 2 columns) in a 3-D space (because of 3 rows).

Note that get.direction() does not provide any possibility to generate a complex direction, such as the 3 main directions of an ellipsoid which has been rotated by the combination of three basic rotations. To understand, you must start with the trdiaional unrotated system with three axes OX, OY and OZ. Then:
- you rotate the system AZ degrees around OZ: this gives the new system OX', OY',OZ'=OZ
- then you rotate this new system AY around the OY' axis: you obtain the new system: OX", OY"=OY', OZ"
- finally you rotate the new system around OX" and obtain the final system.
To my knowledge, these angles are called respectively aximuth, dip and pitch.

To facilitate the calculation of the directions along the main axes of the final system we also provided a tool called
util.ang2mat() where you simply specify the space dimension (i.e. 3) and the list of the three angles.
This directly provides the a 3x3 matrix which can be entered as the 'dirvect' for the definition of the directions: each one of the three columns corresponds to one axis of the final system

Hope this will help.

IF it solved your problem, please modify the name of your POST by adding "[SOLVED]" before its contents.
Didier Renard
 
Posts: 337
Joined: Thu Sep 20, 2012 4:22 pm


Return to Variography

Who is online

Users browsing this forum: No registered users and 2 guests

cron