Skip to content

Move facemelt after calving for mass budgets#142

Open
trhille wants to merge 1 commit into
MALI-Dev:developfrom
trhille:mali/move_facemelt_after_calving
Open

Move facemelt after calving for mass budgets#142
trhille wants to merge 1 commit into
MALI-Dev:developfrom
trhille:mali/move_facemelt_after_calving

Conversation

@trhille
Copy link
Copy Markdown

@trhille trhille commented Mar 27, 2025

This merge moves face-melt after calving to close mass budgets when using RK time integration.

@trhille
Copy link
Copy Markdown
Author

trhille commented Mar 27, 2025

Before this fix:
image

With this fix:
image

@trhille trhille changed the title Mali/move facemelt after calving Move facemelt after calving for mass budgets Mar 27, 2025
@matthewhoffman matthewhoffman self-requested a review July 2, 2025 18:16
@matthewhoffman matthewhoffman self-assigned this Jul 2, 2025
Move face-melt after calving to close mass budgets when using RK time
integration.
@matthewhoffman matthewhoffman force-pushed the mali/move_facemelt_after_calving branch from 8c07da7 to 4f3fc58 Compare July 2, 2025 18:18
@matthewhoffman
Copy link
Copy Markdown

Rebased because the first commit of the previous version of this PR was recently merged to develop in #141 . Rebasing also makes testing easier after recent answer-changing merges.

@matthewhoffman
Copy link
Copy Markdown

Testing

All tests pass except the two with facemelting on, as expected:

landice/humboldt/mesh-3km_restart_test/velo-fo_calving-von_mises_stress_damage-threshold_faceMelting
  * step: full_run
  * step: restart_run
  test execution:      SUCCESS
  test validation:     PASS
  baseline comparison: FAIL
  see: case_outputs/landice_humboldt_mesh-3km_restart_test_velo-fo_calving-von_mises_stress_damage-threshold_faceMelting.log
  test runtime:        00:27
landice/humboldt/mesh-3km_restart_test/velo-fo-depthInt_calving-von_mises_stress_damage-threshold_faceMelting
  * step: full_run
  * step: restart_run
  test execution:      SUCCESS
  test validation:     PASS
  baseline comparison: FAIL
  see: case_outputs/landice_humboldt_mesh-3km_restart_test_velo-fo-depthInt_calving-von_mises_stress_damage-threshold_faceMelting.log
  test runtime:        00:15

Differences are large. Unclear if they are expected to be this large.

thickness            Time index: 0, 1, 2, 3, 4, 5, 6
1:  l1: 4.53476439267530e+02  l2: 3.95597777894097e+02  linf: 3.95324848615629e+02
2:  l1: 5.01116353699082e+02  l2: 4.07109970964872e+02  linf: 4.06335759318895e+02
3:  l1: 6.26132505228975e+02  l2: 4.27198603333262e+02  linf: 4.23650439317753e+02
4:  l1: 5.92921539262783e+02  l2: 3.69121481351878e+02  linf: 3.63436087243054e+02
5:  l1: 3.72693581778073e+02  l2: 1.53352865240742e+02  linf: 1.37498987113333e+02
6:  l1: 1.22055590291217e+03  l2: 5.13559180111984e+02  linf: 2.99043440218658e+02
  FAIL /pscratch/sd/h/hoffman2/COMPASS/TESTING/landice/humboldt/mesh-3km_restart_test/velo-fo_calving-von_mises_stress_damage-threshold_faceMelting/full_run/output.nc
       /pscratch/sd/h/hoffman2/COMPASS/BASELINE/landice/humboldt/mesh-3km_restart_test/velo-fo_calving-von_mises_stress_damage-threshold_faceMelting/full_run/output.nc
surfaceSpeed         Time index: 0, 1, 2, 3, 4, 5, 6
1:  l1: 2.79210014063515e-05  l2: 3.93366526411274e-06  linf: 1.22763442527712e-06
2:  l1: 1.24748301578527e-04  l2: 3.58770963886290e-05  linf: 1.94480802450059e-05
3:  l1: 9.31764772482345e-05  l2: 2.13331150159263e-05  linf: 1.34009811939723e-05
4:  l1: 1.12710416073731e-04  l2: 2.97791198903100e-05  linf: 1.86806316848164e-05
5:  l1: 5.00904093553387e-05  l2: 1.54888524885740e-05  linf: 1.41361007631946e-05
6:  l1: 1.92962371265009e-04  l2: 4.59135343037054e-05  linf: 1.99461232239381e-05
  FAIL /pscratch/sd/h/hoffman2/COMPASS/TESTING/landice/humboldt/mesh-3km_restart_test/velo-fo_calving-von_mises_stress_damage-threshold_faceMelting/full_run/output.nc
       /pscratch/sd/h/hoffman2/COMPASS/BASELINE/landice/humboldt/mesh-3km_restart_test/velo-fo_calving-von_mises_stress_damage-threshold_faceMelting/full_run/output.nc
calvingVelocity      Time index: 0, 1, 2, 3, 4, 5, 6
1:  l1: 2.27136648871665e-06  l2: 1.37510250704195e-06  linf: 1.14344712684904e-06
2:  l1: 6.76039191693177e-06  l2: 1.04128774333649e-06  linf: 6.35510118290823e-07
3:  l1: 3.03257304215736e-05  l2: 7.63918143903464e-06  linf: 4.21436811037737e-06
4:  l1: 2.81886329225003e-05  l2: 5.18849018352541e-06  linf: 2.56341113278599e-06
5:  l1: 3.15702284924002e-05  l2: 6.48021580858719e-06  linf: 3.63892217046229e-06
6:  l1: 1.40112898906774e-05  l2: 3.41646934779417e-06  linf: 2.75517528189279e-06
  FAIL /pscratch/sd/h/hoffman2/COMPASS/TESTING/landice/humboldt/mesh-3km_restart_test/velo-fo_calving-von_mises_stress_damage-threshold_faceMelting/full_run/output.nc
       /pscratch/sd/h/hoffman2/COMPASS/BASELINE/landice/humboldt/mesh-3km_restart_test/velo-fo_calving-von_mises_stress_damage-threshold_faceMelting/full_run/output.nc
calvingThickness     Time index: 0, 1, 2, 3, 4, 5, 6
1:  l1: 4.15981383737145e+02  l2: 4.03092549432501e+02  linf: 4.02919960073356e+02
2:  l1: 4.48586072816766e+01  l2: 3.06986079271361e+01  linf: 2.86848836117890e+01
3:  l1: 8.72468299474519e+02  l2: 5.81584937240182e+02  linf: 4.32905391349730e+02
4:  l1: 9.22433910651488e+02  l2: 5.53906690700757e+02  linf: 4.17894967186128e+02
5:  l1: 4.36918193485109e+02  l2: 3.57132173798695e+02  linf: 3.53128144241062e+02
6:  l1: 9.63309230889290e+02  l2: 4.90723026710819e+02  linf: 2.94211419110315e+02
  FAIL /pscratch/sd/h/hoffman2/COMPASS/TESTING/landice/humboldt/mesh-3km_restart_test/velo-fo_calving-von_mises_stress_damage-threshold_faceMelting/full_run/output.nc
       /pscratch/sd/h/hoffman2/COMPASS/BASELINE/landice/humboldt/mesh-3km_restart_test/velo-fo_calving-von_mises_stress_damage-threshold_faceMelting/full_run/output.nc
damage               Time index: 0, 1, 2, 3, 4, 5, 6
1:  l1: 1.45664695390704e+00  l2: 9.72352811339513e-01  linf: 7.31514223618498e-01
2:  l1: 1.14959991585932e+00  l2: 7.84719791136013e-01  linf: 6.24145118472409e-01
3:  l1: 3.17427883542075e+00  l2: 1.39670536005383e+00  linf: 7.43609146590579e-01
4:  l1: 2.95493774274052e+00  l2: 1.44685818239264e+00  linf: 8.95091484738934e-01
5:  l1: 1.39777653409225e+00  l2: 9.23628724367822e-01  linf: 6.78600030333255e-01
6:  l1: 2.87769656896924e+00  l2: 1.33667414598130e+00  linf: 7.71894426661928e-01
  FAIL /pscratch/sd/h/hoffman2/COMPASS/TESTING/landice/humboldt/mesh-3km_restart_test/velo-fo_calving-von_mises_stress_damage-threshold_faceMelting/full_run/output.nc
       /pscratch/sd/h/hoffman2/COMPASS/BASELINE/landice/humboldt/mesh-3km_restart_test/velo-fo_calving-von_mises_stress_damage-threshold_faceMelting/full_run/output.nc
faceMeltingThickness Time index: 0, 1, 2, 3, 4, 5, 6
1:  l1: 8.57707987695683e+01  l2: 2.13883973004947e+01  linf: 1.16830454854879e+01
2:  l1: 5.31478287713773e+01  l2: 1.90265972264378e+01  linf: 1.28939274435507e+01
3:  l1: 3.80602117556678e+01  l2: 1.40703352534735e+01  linf: 7.32752441928643e+00
4:  l1: 1.19013383893885e+01  l2: 3.84866881386023e+00  linf: 2.47947088885622e+00
5:  l1: 1.62154046814643e+02  l2: 1.53980923306365e+02  linf: 1.53958780614101e+02
6:  l1: 2.44624755399987e+01  l2: 1.31267727848419e+01  linf: 1.17400271875835e+01
  FAIL /pscratch/sd/h/hoffman2/COMPASS/TESTING/landice/humboldt/mesh-3km_restart_test/velo-fo_calving-von_mises_stress_damage-threshold_faceMelting/full_run/output.nc
       /pscratch/sd/h/hoffman2/COMPASS/BASELINE/landice/humboldt/mesh-3km_restart_test/velo-fo_calving-von_mises_stress_damage-threshold_faceMelting/full_run/output.nc

@trhille
Copy link
Copy Markdown
Author

trhille commented Aug 1, 2025

Note that the change in this PR will break the recently added calving multiplier feature introduced in #137. However, moving face-melting after advection but before calving results in mass budget issues as well when using RK2 (screenshot from Thwaites run below). It's unclear why that's the case.
image

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants