Variogram calculation for point set

[wtrainor]

Hello,

I'm confused about how you would best calculate the experimental variogram of a point set data. The examples I found for your library always seems to generate an exhaustive field first (gs.SRF) which uses an input variogram. If I have (x, y, property), can I just calculate on those? I assume I would do it with a masked array, so I still need to define a grid. Is that correct?

Thank you!
Whitney

[wtrainor]

Nevermind, maybe? I just have to use
mesh_type="unstructured"
in gs.vario_estimate
right?

[LSchueler]

Hello again!
Exactly, in this example we generate a point cloud of values, without a grid. The mesh_type="unstructured" is the default value, that's why you don't see that argument in the example.

[MuellerSeb]

Also the bins can be calculated automatically. So if you got (x, y, property), you can simply call:

import gstools as gs
bin_center, gamma = gs.vario_estimate((x, y), property)

And you get what you want 😊