3 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
4 |
============================== |
============================== |
5 |
|
|
6 |
|
checkpoint47j_post |
7 |
|
o add new exp. (ideal_2D_oce) to test "exotic" parameters and options. |
8 |
|
o moved adexch from forward_step to addummy_in_stepping |
9 |
|
o added exch in ctrl_map_... |
10 |
|
o aveFreq.=0. at end of cost_final to avoid overwrite in |
11 |
|
reverse checkpointing loop |
12 |
|
o Yet more changes in gmredi: |
13 |
|
* adgmredi_calc_tensor |
14 |
|
avoiding all recomputation of gmredi_slope_limit |
15 |
|
* adgmredi_x/y/rtransport |
16 |
|
added flag for excessive storing to avoid recomp. of |
17 |
|
u/v/rtans, dTdx/y/z |
18 |
|
-> this is not really necessary and very memory-consuming |
19 |
|
* adgmredi_slope_psi: |
20 |
|
consistency with gmredi_slope_limit in treatment of GM_slopeSqCutoff |
21 |
|
* gmredi_slope_limit |
22 |
|
re-activated full calculation of taperfct for case 'ac02' |
23 |
|
o updating verification/ adjoints of carbon, global_ocean.90x40x15 |
24 |
|
|
25 |
|
checkpoint47i_post |
26 |
|
o GM_EXCLUDE_() replace GM_TAPER_(): by default, all the code is compiled. |
27 |
|
o add GM_Small_Number & GM_slopeSqCutoff as run-time parameters (data.gmredi) |
28 |
|
default values = 1.D-12 & 1.D+48 (=> give same results as checkpoint47f_post) |
29 |
|
o cg2dTargetResWunit: define tolerance using deltaTmom (and not deltaTfreesurf) |
30 |
|
o gmredi: fix few bugs. |
31 |
|
|
32 |
|
checkpoint47h_post (duplicate tag for PH's c48 tag, in order to move c48) |
33 |
|
This will be a good checkpoint to plug into the ecco-branch. |
34 |
|
o few modif.'s to get latest version adjointed |
35 |
|
(mainly kick out code in ini_linear_phisurf) |
36 |
|
o modif's to run adjoint with exactConserv |
37 |
|
o case GM_BOLUS_ADVEC should be cleaned |
38 |
|
S/R gmredi_slope_psi should be cleaned |
39 |
|
o verification/carbon now has exactConserv=.TRUE. |
40 |
|
|
41 |
|
checkpoint47g_post |
42 |
o time-average AIM output: becomes independent from statvars time-average |
o time-average AIM output: becomes independent from statvars time-average |
43 |
(can be used with #undef ALLOW_TIMEAVE) ; add new variables (e.g., RH) |
(can be used with #undef ALLOW_TIMEAVE) ; add new variables (e.g., RH) |
44 |
|
o After TAF bug fix: |
45 |
|
Tested adjoint of gmredi package. |
46 |
|
Tests were done with checkpoint47 rather than with latest |
47 |
|
checkpoint47f_post since the latter is broken for the adjoint |
48 |
|
verification/carbon/ contains both ADM and TLM gradient checks. |
49 |
|
Here they are, before they get changed/lost/stolen. |
50 |
|
Mostly modif.'s to fix numerical sensitivities. |
51 |
|
Gradient checks OK for |
52 |
|
- GM_taper_scheme: |
53 |
|
* clipping |
54 |
|
* ac02 |
55 |
|
* linear |
56 |
|
* glw91 |
57 |
|
* dm95 |
58 |
|
* ldd97 |
59 |
|
- GMREDI_OPTIONS: |
60 |
|
* GM_VISBECK_VARIABLE_K |
61 |
|
* GM_NON_UNITY_DIAGONAL |
62 |
|
* GM_EXTRA_DIAGONAL |
63 |
|
* GM_BOLUS_ADVEC |
64 |
|
in conjunction with data.gmredi parameters to be checked in |
65 |
|
in a few minutes under verification/carbon/code/ |
66 |
|
|
67 |
checkpoint47f_post |
checkpoint47f_post |
68 |
Merging from release1_p10: |
Merging from release1_p10: |