tx2_3v3 Topography Generation

tx2_3v3 Topography Generation#

This document describes the workflow used to generate the tx2_3v3 ocean model topography and associated files.

Topography Creation Workflow#

1) Create initial topography#

Run:

qsub run_create_topo.tx2_3v3.pbs

This step creates the basic topography file with raw cell-level depth statistics and a first-cut land mask.

Output

topo.sub150.tx2_3v3.SRTM15_V2.4.nc

2) Edit land mask#

Notebook:

MaskEdit_tx2_3v3.ipynb

This notebook edits the land mask, including removing 1-point bays around Greenland and Antarctica.

Output

topo.sub150.tx2_3v3.SRTM15_V2.4.edit1.nc

3) Interpolate and smooth depths#

Run:

qsub run_append_topo_interp_smooth_tx2_3v3.pbs

This step appends a new depth variable D_interp to the topography file with lightly smoothed depths.

4) Apply manual depth edits#

Notebook:

Append_topo_edits_tx2_3v3.ipynb

This notebook applies hand edits to depths for straits and channels.

This step produces the final model input topography.

Final topography variable used by the model:

D_edit2

5) Create channel width file#

Notebook:

Channel_width_tx2_3v3.ipynb

Creates the channel width file used by the model.

Optional Steps (not used in v3)#

6) Channel depth edits for runtime experiments#

Notebook:

Channel_topo_tx2_3v3.ipynb

This notebook allows local topography depth edits for runtime experiments.

7) Incorporate channel edits into the base topography#

If changes from step 6 must be incorporated into the topography generated in step 3, run:

Append_topo_edits.ipynb

Validation / Diagnostic Notebooks#

Several notebooks in the source directory are used to verify each step of the workflow.

CompareFirstMasks.ipynb#

Ensures step 1 reproduces the tx2_3v2 mask.

CompareEditedMasks.ipynb#

Checks the mask after step 2 to confirm that only intended edits were made.

CheckRunoffPoints.ipynb#

Similar validation focused on locations where runoff points were intentionally modified.

CompareDepths.ipynb#

Compares the final depths from step 4 with the input and output files from tx2_3v2.

Expected behavior:

  • Depth differences should appear only at the ~30 points modified for tx2_3v3.

  • Small roundoff differences (O(10⁻³ m)) relative to tx2_3v2 may occur, possibly due to the new grid file.

Contents