| 3 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
| 4 |
============================== |
============================== |
| 5 |
|
|
| 6 |
|
o GCHEM: |
| 7 |
|
- move call to GCHEM_FORCING_SEP, so that it is now called before |
| 8 |
|
the blocking exchanges to avoid computing halos in GCHEM_FORCING_SEP |
| 9 |
|
- introduced a useGCHEM runtime flag and applied it everywhere. |
| 10 |
|
o small cleaning of exch2: remove files that are also in eesupp/src |
| 11 |
|
|
| 12 |
|
checkpoint56 |
| 13 |
|
o AIM: read land-Fraction earlier (from ini-fixed) and monitor land at nIter0 |
| 14 |
|
o mnc: cleanup "full" files--fixes out-of-memory problem |
| 15 |
|
o ECCO-GODAE |
| 16 |
|
* model/, pkg/, code/, input/ now complete w.r.t. ECCO-GODAE production setup |
| 17 |
|
* fully merged and updated from SIO code |
| 18 |
|
windstress sign change affects |
| 19 |
|
- exf_inscal_[u/v]stress |
| 20 |
|
- sign of xx_tau[u/v] adjustments |
| 21 |
|
- signs in cost_scat |
| 22 |
|
* ctrl_get_gen: avoid update of tau[u/v] for first few days |
| 23 |
|
* natl_box_adjoint updated and compressed |
| 24 |
|
* 2-level checkpoint switch added |
| 25 |
|
(tested in EXP. 1 of natl_box_adjoint) |
| 26 |
|
* new switch doInitXX to avoid init. of xx_... in optimcycle=0 |
| 27 |
|
o SVD |
| 28 |
|
* first complete, non-hacked version for SVD calculations |
| 29 |
|
* all interfaces clean w.r.t to handling of xx_..., adxx_..., g_xx_... |
| 30 |
|
performed via direct access files admtlm_vector.it???? |
| 31 |
|
(via ctrl_[un]pack) |
| 32 |
|
* optimcycle is used for ARPACK iteration counter |
| 33 |
|
(note: adxx_... is iterated, rather than xx_...; latter is constant!) |
| 34 |
|
* EV problem ok w.r.t. L2-norm in tangent space; need to check other norms |
| 35 |
|
and add linear transforms where necessary |
| 36 |
|
* controls are initial/final SST, SSS (can be extended; need to check |
| 37 |
|
ctrl_set_[un]pack_xyz.F according to ctrl_set_[un]pack_xy.F |
| 38 |
|
o mdsio: filename info to both stderr, stdout |
| 39 |
|
o exf: exf_inscal_runoff: separated default conversion [m/year] -> [m/sec] |
| 40 |
|
from sign change |
| 41 |
|
|
| 42 |
|
checkpoint55j_post |
| 43 |
|
o isolate dissipation tendency (allow to keep it out off AB) in V.I form. |
| 44 |
|
affects (truncation level) output.txt of global_ocean.cs32x15(&.thsice) |
| 45 |
|
o add consistent build directories and .cvsignore files to verification |
| 46 |
|
o added viscA4GridMin and viscA4GridMax for more control on biharmonic viscosity |
| 47 |
|
o make all subroutines 30 or less characters in length for certain older |
| 48 |
|
compilers to work (a thorough search only turns up two routines in |
| 49 |
|
pkg/ptracers) |
| 50 |
|
o allow (vecinv only) a different horiz. viscosity for Divergence and Vorticity; |
| 51 |
|
affects (truncation level) output.txt of global_ocean.cs32x15(&.thsice) |
| 52 |
|
|
| 53 |
|
checkpoint55i_post |
| 54 |
|
o change arrays for total vertical diffusivity (2.D if explicit, |
| 55 |
|
3.D if implicit) ; each ptracer uses its own background value. |
| 56 |
|
o added mnc/NetCDF output capability to ocn_compon_interf |
| 57 |
|
- added bits to verification/cpl_aim+ocn so that it can run on |
| 58 |
|
ITRDA and other clusters without the mpirun "-wd" option |
| 59 |
|
o fixed wayward endif in mom_fluxform.F that messed up cyclindrical |
| 60 |
|
configuration |
| 61 |
|
o give MNC the ability to create additional files rather than |
| 62 |
|
exceeding a specified file size limit "MNC_MAX_FILESIZE" |
| 63 |
|
- the default is just less than 2GB which is a practical limit |
| 64 |
|
on many systems |
| 65 |
|
o Compute the non-local term using either KPP vertical diffusivity (=default) |
| 66 |
|
or the total vertical diffusivity (when KPP_ghatUseTotalDiffus=T) |
| 67 |
|
|
| 68 |
|
checkpoint55h_post |
| 69 |
|
o add verification test "vermix" |
| 70 |
|
o remove all the *_ioinc flags and replace with the single global |
| 71 |
|
outputTypesInclusive flag |
| 72 |
|
o allow to set a vertical profile of vertical diffusivity for T & S |
| 73 |
|
o call PACKAGE_BOOT & PACKAGES_READPARMS just after INI_PARMS |
| 74 |
|
o useMNC moved from data.mnc to data.pkg |
| 75 |
|
|
| 76 |
|
checkpoint55g_post |
| 77 |
|
o Renaming carbon AD experiments to comply with testreport conventions |
| 78 |
|
o Updating adjoint of Held-Suarez hs94.1x64x5 |
| 79 |
|
o Fixing directives for non-ECCO forward_step, after ECCO directives |
| 80 |
|
have been fixed to avoid recomp. of do_oceanic_phys |
| 81 |
|
Works well for global_ocean.90x40x15, |
| 82 |
|
but recomp. issue remains for carbon |
| 83 |
|
o Updating adjoint output for natl_box_adjoint after |
| 84 |
|
pkg/ecco has been modified (see taga c55d_post) |
| 85 |
|
|
| 86 |
|
checkpoint55f_post |
| 87 |
|
o rename to timeave_init_fixed |
| 88 |
|
|
| 89 |
|
checkpoint55e_post |
| 90 |
|
o some delicate re-shuffle of store directives to avoid one |
| 91 |
|
extra call of do_oceanic_physics |
| 92 |
|
affects ecco_the_main_loop.F, do_oceanic_physics.F |
| 93 |
|
checkpoint_lev1_directives.h |
| 94 |
|
o NB: this may break global_ocean adjoint temporarily, |
| 95 |
|
but it is clear how to fix it. Will do later, need this now. |
| 96 |
|
|
| 97 |
checkpoint55d_post |
checkpoint55d_post |
| 98 |
o ECCO specific cost function terms (up-to-date with 1x1 runs) |
o ECCO specific cost function terms (up-to-date with 1x1 runs) |
| 99 |
o ecco_cost_weights is modified to 1x1 runs |
o ecco_cost_weights is modified to 1x1 runs |
Parent Directory
| 
Revision Log
| 
Revision Graph
| 
Patch