Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add new IcoswISC30E3r2 ocean and sea-ice mesh #5931

Closed
wants to merge 11 commits into from

Conversation

xylar
Copy link
Contributor

@xylar xylar commented Sep 18, 2023

Long name: IcoswISC30L64E3SMv3r2

This Icosahedral (Icos) mesh is made by subdividing an icosahedron and has nearly uniform 30 km resolution.

This mesh is a candidate for the E3SM v3 (E3) low res mesh. It is revision 2 (r2) of the mesh. (Revision 1 had incorrect bathymetry and coastline around Greenland.)

The mesh was created using compass, specifically this tag: https://github.com/MPAS-Dev/compass/releases/tag/mesh_IcoswISC30E3r2

A G-case with data iceberg melt fluxes (DIB) and data ice-shelf melt fluxes (DISMF) has been run for 30 years, with MPAS-Analysis output available here:
https://web.lcrc.anl.gov/public/e3sm/diagnostic_output/ac.xylar/analysis/20230905.G-test.IcoswISC30E3r2.anvil/

The mesh and the G-case results is still been reviewed and approved here:
https://acme-climate.atlassian.net/wiki/spaces/OO/pages/3904831507/Review+IcoswISC30E3r2

A B-case will begin shortly and analysis will be posted here and on the review page as soon as it is available.

@xylar xylar added mpas-ocean mpas-seaice BFB PR leaves answers BFB labels Sep 18, 2023
@xylar xylar changed the title Add new ECwISC30to60E3r2 ocean and sea-ice mesh Add new IcoswISC30E3r2 ocean and sea-ice mesh Sep 18, 2023
@xylar
Copy link
Contributor Author

xylar commented Sep 18, 2023

This branch is based off of #5927 and will be rebased after that PR is merged.

@xylar
Copy link
Contributor Author

xylar commented Sep 18, 2023

Relevant discussion at E3SM-Ocean-Discussion#64

@xylar
Copy link
Contributor Author

xylar commented Sep 18, 2023

@jonbob, what about these files? Should I sync over the inputdata and diagnostics files for these, too?

Copy link
Contributor

@mark-petersen mark-petersen left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed mesh, coastlines, critical passages, and MPAS-Analysis simulation results.

Copy link
Contributor

@proteanplanet proteanplanet left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Through-flow, shallow bathymetry, cavity geometry and coastlines have been checked.

Copy link
Contributor

@erinethomas erinethomas left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

approved - looked at MPAS-analysis and mesh.

<gridmap atm_grid="ne30np4.pg2" ocn_grid="ECwISC30to60E3r2">
<map name="ATM2OCN_FMAPNAME">cpl/gridmaps/ne30pg2/map_ne30pg2_to_ECwISC30to60E3r2_mono.230901.nc</map>
<map name="ATM2OCN_VMAPNAME">cpl/gridmaps/ne30pg2/map_ne30pg2_to_ECwISC30to60E3r2_bilin.230901.nc</map>
<map name="ATM2OCN_SMAPNAME">cpl/gridmaps/ne30pg2/map_ne30pg2_to_ECwISC30to60E3r2-nomask_bilin.230901.nc</map>
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Was this made with TempestRemap?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@jonbob, that's a question for you.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We've been making all conservative "mono" mapping files with tempest, but not the bilinear ones. I'll try to change the process to match the new naming convention and process

@jonbob
Copy link
Contributor

jonbob commented Sep 26, 2023

20-year B-case using the CRYO1850 compset has completed:
MPAS-Analysis for: 20230918.test5931.anvil

@rljacob rljacob added this to the v3.0alpha05 milestone Oct 12, 2023
Copy link
Member

@rljacob rljacob left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Bilinear maps need to be remade with TR. Map files need to be renamed to follow naming convention.

@jonbob
Copy link
Contributor

jonbob commented Nov 14, 2023

The new mapping and domain files have been staged to /lcrc/group/e3sm/data/inputdata and have been tested successfully with:

  • SMS_Ld3.T62_IcoswISC30E3r2.CMPASO-IAF.chrysalis_intel
  • SMS_Ld3.TL319_IcoswISC30E3r2.GMPAS-JRA1p5.chrysalis_intel
  • SMS_Ld3.ne30pg2_IcoswISC30E3r2.WCYCL1850.chrysalis_intel
  • SMS_Ld3.ne30pg2_r05_IcoswISC30E3r2.WCYCL1850.chrysalis_intel
  • SMS_Ld3.ne30pg2_r05_IcoswISC30E3r2.WCYCL1850.chrysalis_intel with trfvnp2 mapping file

@jonbob
Copy link
Contributor

jonbob commented Nov 14, 2023

@rljacob -- can you check that the mapping files are now consistent with the algorithm and naming conventions?

@xylar xylar force-pushed the mesh/add-icoswisc30e3r2 branch from 701f92c to f270208 Compare November 14, 2023 20:16
cime_config/config_grids.xml Outdated Show resolved Hide resolved
@xylar
Copy link
Contributor Author

xylar commented Nov 14, 2023

@jonbob, please verify that the grid files look correct after the rebase. I don't expect any problems.

@rljacob, could you verify that your concerns have been addressed by @jonbob's latest commit?

@xylar
Copy link
Contributor Author

xylar commented Nov 14, 2023

@alicebarthel, could you either approve or let us know if you have concerns about this mesh?

@jonbob
Copy link
Contributor

jonbob commented Nov 14, 2023

@xylar -- thanks for rebasing. The grid files looked good to me from a visual inspection, and I ran a quick ne30pg2_r05_IcoswISC30E3r2.WCYCL1850 smoke test to check -- and it has been successful as well

<gridmap atm_grid="T62" ocn_grid="IcoswISC30E3r2">
<map name="ATM2OCN_FMAPNAME">cpl/gridmaps/T62/map_T62_to_IcoswISC30E3r2_traave.20231113.nc</map>
<map name="ATM2OCN_SMAPNAME">cpl/gridmaps/T62/map_T62_to_IcoswISC30E3r2-nomask_trbilin.20231113.nc</map>
<map name="ATM2OCN_VMAPNAME">cpl/gridmaps/T62/map_T62_to_IcoswISC30E3r2_patch.20231113.nc</map>
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

"patch" should be "esmfpatch"

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It should be fixed now

<gridmap atm_grid="TL319" ocn_grid="IcoswISC30E3r2">
<map name="ATM2OCN_FMAPNAME">cpl/gridmaps/TL319/map_TL319_to_IcoswISC30E3r2_traave.20231113.nc</map>
<map name="ATM2OCN_SMAPNAME">cpl/gridmaps/TL319/map_TL319_to_IcoswISC30E3r2-nomask_trbilin.20231113.nc</map>
<map name="ATM2OCN_VMAPNAME">cpl/gridmaps/TL319/map_TL319_to_IcoswISC30E3r2_patch.20231113.nc</map>
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

"patch" should be "esmfpatch"

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

should be fixed now

@@ -4540,6 +4638,11 @@
<map name="ROF2OCN_LIQ_RMAPNAME">cpl/cpl6/map_rx1_to_ECwISC30to60E3r2_smoothed.r150e300.230901.nc</map>
</gridmap>

<gridmap ocn_grid="IcoswISC30E3r2" rof_grid="rx1">
<map name="ROF2OCN_ICE_RMAPNAME">cpl/cpl6/map_rx1_to_IcoswISC30E3r2_smoothed.r150e300.230905.nc</map>
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

"smoothed" should be "cstmnn" for "custom nearest neighbor".

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

should be fixed now -- I also changed the datestamp to YYYYMMDD to match the new naming convention

@@ -4540,6 +4638,11 @@
<map name="ROF2OCN_LIQ_RMAPNAME">cpl/cpl6/map_rx1_to_ECwISC30to60E3r2_smoothed.r150e300.230901.nc</map>
</gridmap>

<gridmap ocn_grid="IcoswISC30E3r2" rof_grid="rx1">
<map name="ROF2OCN_ICE_RMAPNAME">cpl/cpl6/map_rx1_to_IcoswISC30E3r2_smoothed.r150e300.230905.nc</map>
<map name="ROF2OCN_LIQ_RMAPNAME">cpl/cpl6/map_rx1_to_IcoswISC30E3r2_smoothed.r150e300.230905.nc</map>
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

"smoothed" should be "cstmnn"

@@ -4600,6 +4703,11 @@
<map name="ROF2OCN_LIQ_RMAPNAME">cpl/cpl6/map_JRA025_to_ECwISC30to60E3r2_smoothed.r150e300.230901.nc</map>
</gridmap>

<gridmap ocn_grid="IcoswISC30E3r2" rof_grid="JRA025">
<map name="ROF2OCN_ICE_RMAPNAME">cpl/cpl6/map_JRA025_to_IcoswISC30E3r2_smoothed.r150e300.230905.nc</map>
<map name="ROF2OCN_LIQ_RMAPNAME">cpl/cpl6/map_JRA025_to_IcoswISC30E3r2_smoothed.r150e300.230905.nc</map>
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

smoothed should be "cstmnn" for both.

@@ -4685,6 +4793,11 @@
<map name="ROF2OCN_LIQ_RMAPNAME">cpl/cpl6/map_r05_to_ECwISC30to60E3r2_smoothed.r150e300.230901.nc</map>
</gridmap>

<gridmap ocn_grid="IcoswISC30E3r2" rof_grid="r05">
<map name="ROF2OCN_ICE_RMAPNAME">cpl/cpl6/map_r05_to_IcoswISC30E3r2_smoothed.r150e300.230905.nc</map>
<map name="ROF2OCN_LIQ_RMAPNAME">cpl/cpl6/map_r05_to_IcoswISC30E3r2_smoothed.r150e300.230905.nc</map>
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

"smoothed" should be "cstmnn" for both.

@rljacob
Copy link
Member

rljacob commented Nov 15, 2023

The r2o maps need a name fix but the rest look good. Also make sure these are on the input data server.

@@ -476,6 +491,7 @@
<config_btr_dt ocn_grid="SOwISC12to60E2r4">'0000_00:00:15'</config_btr_dt>
<config_btr_dt ocn_grid="ECwISC30to60E2r1">'0000_00:01:00'</config_btr_dt>
<config_btr_dt ocn_grid="ECwISC30to60E3r2">'0000_00:01:00'</config_btr_dt>
<config_btr_dt ocn_grid="IcoswISC30E3r2">'0000_00:00:45'</config_btr_dt>
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

(documenting a slack conversation) This icos ~30km should be able to run with exactly the same time step settings as the EC30to60, since it only depends on the minimum cell width. That means

<config_dt ocn_grid="IcoswISC30E3r2">'00:30:00'</config_dt>  (it's already set to that)
<config_btr_dt ocn_grid="IcoswISC30E3r2">'0000_00:01:15'</config_btr_dt>

The 45s was overly cautious to begin with. 1:15 should work.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@mark-petersen -- in testing 1:15 caused mpaso state validation errors, but 1:00 has been stable

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

OK, thanks. The barotropic time step is limited by an advective CFL condition for the surface gravity waves. There must be some faster velocities on the ICOS30 in locations with 30km cells, where the EC30to60 has larger cells (like mid-latitude western boundary currents). Regardless, we always set this by trial and error in the end. Thanks for testing it. The 1:00 is still a big improvement over 45s.

@jonbob
Copy link
Contributor

jonbob commented Nov 16, 2023

@rljacob -- I think I fixed all the naming issues, and changed the filenames on the lcrc local inputdata location. Once you approve, I'll copy them to the inputdata server

Copy link
Contributor

@alicebarthel alicebarthel left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I approve based on visual inspection of the mesh plots and the G- and B-case MPAS-Analysis plots. Some enhanced biases due to coupling but nothing that look outside of our normal range of issues, or would warrant blocking this configuration in my opinion. Well done!

@xylar xylar marked this pull request as draft November 20, 2023 00:53
@xylar
Copy link
Contributor Author

xylar commented Nov 20, 2023

I'm putting this in draft mode because I just discovered a bug in landIceMask (which affects the initial condition as well as several mapping files) over the weekend.

MPAS-Dev/compass#732

@xylar
Copy link
Contributor Author

xylar commented Nov 20, 2023

I'm closing this and will make a new PR for IcoswISC30E3r4 with the bug fix and topography smoothing.

@xylar xylar closed this Nov 20, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
BFB PR leaves answers BFB mpas-ocean mpas-seaice
Projects
None yet
Development

Successfully merging this pull request may close these issues.

8 participants