Spatialreference or equivilent for Disv-grid (mf6)

[DStrom1987]

Hi,

I've been struggling a bit with the pstfrom module since it needs information about the grid dimensions by use of the function SpatialReference.

As this functionality revolves around a structured x, y-grid approach I cant come up with a way to set it up for my unstructured models.

I have looked into some older threads on the topic where there are suggestions of an approach;
#246

pyemu/autotest/pst_from_tests.py

Line 2840 in e1ce35a

def usg_freyberg_test():

However, it feels a bit overcomplicated, especially for me whos not really fluent in python (or programming for that matter).

Since the thread is from 2021, I was wondering if anyone has come up with a more elegant and ready to go solution since then?

[jtwhite79]

We are actively working on pypestutils (https://github.com/pypest/pypestutils) and it has native support for unstructured grids in realm of both parameters and model post processing (like mod2obs, etc). Its not quite ready-to-go, but Ive got a few examples and some helpers setup on my fork (will be merged into the main repo soon) that work on a quadtree variant of the Freyeberg model and it seems to be working.

[DStrom1987]

Great, thanks for the info. Looks like a very solid option for what I want to achieve. I'll look into it.

//David

[DStrom1987]

I ran some tests on my data with pypestutils but I'm running into a problem...

As the image shows, I'm facing issues when flopy tries to load the disv-file. The file in my model is generated through modelmuse rather than flopy. But looking at them (attatched) I think they follow the same structure. The only difference being how layers are considerd as separate datasets in the pypestutils-example.

I'll attatch both disv-files to show how they are structured

freyberg:

`# file generated by flopy version 3.3.6 on 12/30/2022 at 13:26:49.
begin options
length_units meters
angrot -55
end options

begin dimensions
nlay 3
ncpl 5108
nvert 6196
end dimensions

begin griddata
top
open/close 'freyberg6.disv_top.txt' factor 1.0
botm layered
open/close 'freyberg6.disv_botm_layer1.txt' factor 1.0
open/close 'freyberg6.disv_botm_layer2.txt' factor 1.0
open/close 'freyberg6.disv_botm_layer3.txt' factor 1.0
idomain layered
open/close 'freyberg6.disv_idomain_layer1.txt' factor 1
open/close 'freyberg6.disv_idomain_layer2.txt' factor 1
open/close 'freyberg6.disv_idomain_layer3.txt' factor 1
end griddata

begin vertices
open/close 'freyberg6.disv_vertices.txt'
end vertices

begin cell2d
open/close 'freyberg6.disv_cell2d.txt'

end cell2d
`

my example-model

Attatched as a csv to fit the message.
mod0000.csv

[jtwhite79]

@DStrom1987 Im not sure what is happening with that disv stuff - Im not an unstructured grid person. But pypestuils can get grid info it needs from the MODFLOW-6 binary grid file (grb) (assuming you set the xorigin, yorigin, and angrot options in the disv package)...so you dont need flopy to use pypestutils...feel free to open issues and discussions on pypestutils as well!

[DStrom1987]

Thanks, I'm kind of a python newbie, so I really still need examples to work with. Atleast it makes things much easier. Which function would i use in pypestutils for this operation and how would I call it? Sincerely David Den ons 25 okt. 2023 21:57J Dub ***@***.***> skrev:

…

@DStrom1987 <https://github.com/DStrom1987> Im sure what is happening with that disv stuff - Im not an unstructured grid person. But pypestuils can get grid info it needs from the MODFLOW-6 binary grid file (grb) (assuming you set the xorigin, yorigin, and angrot options in the disv package)...so you dont need flopy to use pypestutils...feel free to open issues and discussions on pypestutils as well! — Reply to this email directly, view it on GitHub <#467 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ARYO5IEVITI7IIIAIVQVRR3YBFVKXAVCNFSM6AAAAAA6MUBNDKVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM3TGOBVGEYTE> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[jtwhite79]

The develop branch of pypestuils has some examples. Admittedly, we need to make a new release so that the latest version of pypestutils is available to work with these examples (unless you are ok building the library yourself? - build instructions). Otherwise, we are planning a release in the next couple of days...stay tuned!

[DStrom1987]

Of course! Perhaps that's why the functions I'm calling are not recognized. I think I'll wait for the release since I dont have so much time on my hands in the coming days anyway. Sincerely David Den ons 25 okt. 2023 kl 22:53 skrev J Dub ***@***.***>:

…

The develop branch of pypestuils has some examples <https://github.com/pypest/pypestutils/tree/develop/examples>. Admittedly, we need to make a new release so that the latest version of pypestutils is available to work with these examples (unless you are ok building the library yourself? - build instructions <https://github.com/pypest/pypestutils/blob/main/BUILD.md>). Otherwise, we are planning a release in the next couple of days...stay tuned! — Reply to this email directly, view it on GitHub <#467 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ARYO5IFFVHRL5U5MNXWT2TTYBF34VAVCNFSM6AAAAAA6MUBNDKVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM3TGOBVG4YDG> . You are receiving this because you were mentioned.Message ID: ***@***.***>