[vincentpost]

This version introduces layer model version 3.0.0 which includes the new aquitard S2.1A. Moreover changes have been made to botm.py and scripts were introduced in the dikte_aquitard folder to create a mask for S2.1A, to create boundaries for each layer based on the convex hull of the data points and to merge the masks (which was done manually in QGIS before).

[bdestombe]

botm.geojson Still uses the old koppeltabel layering scheme

[bdestombe]

./conductances should also be edited. The conductances of the new S21A—Kreftenheye Formation should be introduced and subtracted from another layer. Both S13 as S21 have a c value of 2000 days and have non-zero thickness at the location where S21A should get valid c-values, so we will have to have a look at the source material of Stuyfzand. @vincentpost

[bdestombe]

I just committed the changes to the conductance folder. Basically I used the conductance values from the Bergen model layer 1D: 1000 days everywhere.

[bdestombe]

And we where struggling with botm data points, from which the locations were also used to setup the mask. If I remember correctly, we chose to relocate the datapoints slightly so they are not masked.

What if we make a change to NHFLO/tools that all data points are used, and the mask is applied after interpolating the bottoms to the modflow grid. The interpolated grid will be slightly different. The main advantage here is that we don't have to shift the data points.

[bdestombe]

Next time, smaller commits. Now it is hard to distinguish the difference between just copying v2.0.0 and your own changes.

[bdestombe]

# Add src directory to path so nhflodata module can be found
src_path = Path(__file__).resolve().parents[6]
if str(src_path) not in sys.path:
    sys.path.insert(0, str(src_path))

Als je de repo installeert in je environment met pip install -e ., kan je de functie gewoon draaien en pakt hij met get_abs_data_path() de databestanden uit de mockup folder.

[bdestombe]

Remaining review questions and follow-up work are tracked in #78: exported bottom-layer naming consistency, S21A resistance allocation, interpolation before masking, and remaining script import cleanup. The issue distinguishes these from changes already implemented in this PR and links the original discussion. Merging this PR does not close the follow-up issue.