| 4 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
| 5 |
============================== |
============================== |
| 6 |
|
|
| 7 |
|
o pkg/seaice: simplify the use of CPP flags in JFNK-related files |
| 8 |
|
(compile when SEAICE_ALLOW_JFNK is defined) |
| 9 |
|
o verification/global_ocean.90x40x15 and OpenAD: |
| 10 |
|
New experiment for OpenAD is identical to the one in input_ad except that |
| 11 |
|
multiDimAdvection=.FALSE., |
| 12 |
|
There is still an issue with this. |
| 13 |
|
o pkg/thsice: |
| 14 |
|
- account for ice velocity (ocean relative velocity) in turbulent heat flux |
| 15 |
|
at the base of the ice; update output of global_ocean.cs32x15 (icedyn) |
| 16 |
|
and offline_exf_seaice (main test). |
| 17 |
|
|
| 18 |
|
checkpoint64f (2013/03/31) |
| 19 |
|
o verification/OpenAD: |
| 20 |
|
- new (smaller, without OAD_DEBUG) output_oadm.txt generated using gfortran. |
| 21 |
|
o verification/isomip, pkg/shelfice and OpenAD: |
| 22 |
|
- add OpenAD setup to this exp and make shelfice package work with OpenAD |
| 23 |
|
o verification/tutorial_global_oce_biogeo: |
| 24 |
|
- update AD and TLM output after turning off ALLOW_FE (in DIC_OPTIONS.h) |
| 25 |
|
o verification/offline_exf_seaice: |
| 26 |
|
- replace adjoint experiment (now similar to fwd input.thermo/ setup). |
| 27 |
|
o verification/tutorial_tracer_adjsens, pkg/ptracers and OpenAD: |
| 28 |
|
- add OpenAD setup to this exp and make ptracers package work with OpenAD |
| 29 |
|
o pkg/seaice/seaice_evp.F: |
| 30 |
|
mask the asymmetric contribution of the ice-ocean stress (when |
| 31 |
|
the turningAngle is non-zero) over open water to avoid stripes |
| 32 |
|
in the EVP-solutions after many EVP-substeps. This does not change |
| 33 |
|
the verification experiment (because turningAngle=0 in that experiment), |
| 34 |
|
but will affect other EVP solutions with turningAngle.ne.0 |
| 35 |
|
o verification, AD & TLM output: |
| 36 |
|
- new TAF version 2.3.10 (previously 2.3.8) add more parenthesis in multiple |
| 37 |
|
additions and multiplications which affect results at machine truncation |
| 38 |
|
level. Update few AD output (1D_ocean_ice_column, bottom_ctrl_5x5, |
| 39 |
|
lab_sea.noseaicedyn & offline_exf_seaice) and TLM output |
| 40 |
|
(1D_ocean_ice_column & tutorial_tracer_adjsens.som81) which had only 12 |
| 41 |
|
digits of matching. |
| 42 |
|
o pkg/seaice: |
| 43 |
|
modify lsr-algorithm so that it vectorizes AND converges by using a |
| 44 |
|
"zebra" method: split vector loop into two with step size 2; this |
| 45 |
|
solves a general convergence problem for the JFNK solver on a vector |
| 46 |
|
computer. It has the potential to even improve convergence for |
| 47 |
|
non-vector machines, but it does change the results so it is turned |
| 48 |
|
off by default. To turn on define CPP-flags SEAICE_VECTORIZE_LSR and |
| 49 |
|
SEAICE_VECTORIZE_LSR_ZEBRA in SEAICE_OPTIONS.h |
| 50 |
|
|
| 51 |
checkpoint64e (2013/03/05) |
checkpoint64e (2013/03/05) |
| 52 |
o pkg/ecco and several verification experiments: |
o pkg/ecco and several verification experiments: |
| 53 |
Remove special versions of the_model_main.F, the_main_loop.F, forward_step.F |
Remove special versions of the_model_main.F, the_main_loop.F, forward_step.F |
Parent Directory
| 
Revision Log
| 
Revision Graph
| 
Patch