3 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
4 |
============================== |
============================== |
5 |
|
|
6 |
|
o r* coordinate added in #ifdef NONLIN_FRSURF block. |
7 |
|
(modification to pressure gradient not yet implemented) |
8 |
|
o mom_fluxform: new S/R mom_calc_rtrans.F computes vertical transports |
9 |
|
(for U & V) that are used for vertical advection of momentum. |
10 |
|
|
11 |
|
checkpoint48a_post |
12 |
|
o In preparation for r*, use temp. array to compute vertical transport |
13 |
|
in flux-form momentum pkg. => change results of few test-exp. |
14 |
|
Update output of: exp4, exp5, front_relax, global_ocean_pressure, |
15 |
|
lab_sea, natl_box & plume_on_slope |
16 |
|
|
17 |
|
checkpoint48 |
18 |
|
o fix small bug that was causing lot of warning (TIMER_STOP 'UPDATE_CG2D') |
19 |
|
|
20 |
|
checkpoint47j_post |
21 |
|
o add new exp. (ideal_2D_oce) to test "exotic" parameters and options. |
22 |
|
o moved adexch from forward_step to addummy_in_stepping |
23 |
|
o added exch in ctrl_map_... |
24 |
|
o aveFreq.=0. at end of cost_final to avoid overwrite in |
25 |
|
reverse checkpointing loop |
26 |
|
o Yet more changes in gmredi: |
27 |
|
* adgmredi_calc_tensor |
28 |
|
avoiding all recomputation of gmredi_slope_limit |
29 |
|
* adgmredi_x/y/rtransport |
30 |
|
added flag for excessive storing to avoid recomp. of |
31 |
|
u/v/rtans, dTdx/y/z |
32 |
|
-> this is not really necessary and very memory-consuming |
33 |
|
* adgmredi_slope_psi: |
34 |
|
consistency with gmredi_slope_limit in treatment of GM_slopeSqCutoff |
35 |
|
* gmredi_slope_limit |
36 |
|
re-activated full calculation of taperfct for case 'ac02' |
37 |
|
o updating verification/ adjoints of carbon, global_ocean.90x40x15 |
38 |
|
|
39 |
|
checkpoint47i_post |
40 |
|
o GM_EXCLUDE_() replace GM_TAPER_(): by default, all the code is compiled. |
41 |
|
o add GM_Small_Number & GM_slopeSqCutoff as run-time parameters (data.gmredi) |
42 |
|
default values = 1.D-12 & 1.D+48 (=> give same results as checkpoint47f_post) |
43 |
|
o cg2dTargetResWunit: define tolerance using deltaTmom (and not deltaTfreesurf) |
44 |
|
o gmredi: fix few bugs. |
45 |
|
|
46 |
|
checkpoint47h_post (duplicate tag for PH's c48 tag, in order to move c48) |
47 |
|
This will be a good checkpoint to plug into the ecco-branch. |
48 |
|
o few modif.'s to get latest version adjointed |
49 |
|
(mainly kick out code in ini_linear_phisurf) |
50 |
|
o modif's to run adjoint with exactConserv |
51 |
|
o case GM_BOLUS_ADVEC should be cleaned |
52 |
|
S/R gmredi_slope_psi should be cleaned |
53 |
|
o verification/carbon now has exactConserv=.TRUE. |
54 |
|
|
55 |
|
checkpoint47g_post |
56 |
|
o time-average AIM output: becomes independent from statvars time-average |
57 |
|
(can be used with #undef ALLOW_TIMEAVE) ; add new variables (e.g., RH) |
58 |
|
o After TAF bug fix: |
59 |
|
Tested adjoint of gmredi package. |
60 |
|
Tests were done with checkpoint47 rather than with latest |
61 |
|
checkpoint47f_post since the latter is broken for the adjoint |
62 |
|
verification/carbon/ contains both ADM and TLM gradient checks. |
63 |
|
Here they are, before they get changed/lost/stolen. |
64 |
|
Mostly modif.'s to fix numerical sensitivities. |
65 |
|
Gradient checks OK for |
66 |
|
- GM_taper_scheme: |
67 |
|
* clipping |
68 |
|
* ac02 |
69 |
|
* linear |
70 |
|
* glw91 |
71 |
|
* dm95 |
72 |
|
* ldd97 |
73 |
|
- GMREDI_OPTIONS: |
74 |
|
* GM_VISBECK_VARIABLE_K |
75 |
|
* GM_NON_UNITY_DIAGONAL |
76 |
|
* GM_EXTRA_DIAGONAL |
77 |
|
* GM_BOLUS_ADVEC |
78 |
|
in conjunction with data.gmredi parameters to be checked in |
79 |
|
in a few minutes under verification/carbon/code/ |
80 |
|
|
81 |
|
checkpoint47f_post |
82 |
|
Merging from release1_p10: |
83 |
|
o modifications for using pkg/exf with pkg/seaice |
84 |
|
- pkg/seaice CPP options SEAICE_EXTERNAL_FORCING |
85 |
|
and SEAICE_EXTERNAL_FLUXES |
86 |
|
- pkg/exf CPP options EXF_READ_EVAP and |
87 |
|
EXF_NO_BULK_COMPUTATIONS |
88 |
|
- usage examples are Experiments 8 and 9 in |
89 |
|
verification/lab_sea/README |
90 |
|
- verification/lab_sea default experiment now uses |
91 |
|
pkg/gmredi, pkg/kpp, pkg/seaice, and pkg/exf |
92 |
|
|
93 |
|
checkpoint47e_post |
94 |
|
o created a tag prior to creating a branch |
95 |
|
|
96 |
|
checkpoint47d_post |
97 |
|
o fixed bug in the oceanic pressure coordinates code: vertical viscosity |
98 |
|
at the bottom boundary had an erroneous half slip boundary condition |
99 |
|
o added bottom drag and no slip boundary condition capabilities to |
100 |
|
oceanic pressure coordinates code |
101 |
|
o added new verification experiment global_ocean_pressure, this experiment is |
102 |
|
also described in the tutorial section of the manual |
103 |
|
checkpoint47d_pre |
104 |
|
|
105 |
|
checkpoint47c_post |
106 |
|
o Added a verification experiment for a global cubed-sphere ocean run |
107 |
|
- verification/global_ocean.cubed32x32x30 |
108 |
|
- uses pkg/therm_seaice and pkg/bulk_forcing |
109 |
|
|
110 |
|
o allow to run AIM physics with SPEEDY input files (from Franco Molteni) |
111 |
|
o allow a more accurate definition of Ro_Surf (selectFindRoSurf=1) |
112 |
|
when using P-coordinate; only implemented for atmospheric config. |
113 |
|
o OCEANICP & realFreshWater: include P-E direct effect on wVel ; |
114 |
|
NOTES: requires option NONLIN_FRSURF to be "#define". |
115 |
|
o update advect_xz/results/output.txt (left from checkpoint44g_post) |
116 |
|
|
117 |
|
checkpoint47b_post |
118 |
|
Merging from release1_p9: |
119 |
|
o pkg/seaice |
120 |
|
- removed GOTO's and added taf directives |
121 |
|
- double precision constants to reduce the g77 (Linux) |
122 |
|
to F77 (SGI) differences reported in release1_p8 |
123 |
|
o tools/genmake |
124 |
|
- added SGI options |
125 |
|
o verification/testscript |
126 |
|
- updated to that of checkpoint47a_post |
127 |
|
o verification/global_ocean.90x40x15/input/eedata |
128 |
|
- modified for SGI f77 compatibility |
129 |
|
o verification/lab_sea |
130 |
|
- added description of sea-ice model |
131 |
|
- added missing matlab routines |
132 |
|
- added test of thermodynamics parallelization |
133 |
|
|
134 |
|
checkpoint47a_post |
135 |
|
o new pkg: aim_v23 |
136 |
|
= F. Molteni atmos.physics (SPEEDY, ver23) adapted to MITgcm. |
137 |
|
for now, keep the same surface forcing as before. |
138 |
|
- Part-Cell implemented into AIM; check that Heat & Water are conserved |
139 |
|
- aim.5l_cs: use new aim pkg (run 10yr & get better results than before) |
140 |
|
- aim.5l_LatLon & aim.5l_Equatorial_Channel are still using old aim pkg |
141 |
|
o add diagnostic of surface correction term in monitor output |
142 |
|
|
143 |
|
o bulk_force and therm_seaice |
144 |
|
- Two packages: bulk_force (Bulk forcing) |
145 |
|
and therm_seaice (thermodynamic_seaice) - adopted from LANL CICE.v2.0.2 |
146 |
|
- Earlier integration from Stephaine Dutkiewicz |
147 |
|
and Patrick Heimbach. |
148 |
|
- Two ifdef statements for compile time, |
149 |
|
ALLOW_THERM_SEAICE and ALLOW_BULK_FORCE |
150 |
|
|
151 |
|
checkpoint47 |
152 |
|
differentiable version of checkpoint46n_post |
153 |
|
o bug fix in quasihydrostaticterms |
154 |
|
o * "clean" adjoint code (in terms of extensive recomputations) |
155 |
|
can now be obtained for all GMREDI options (i.e. for |
156 |
|
- GM_VISBECK_VARIABLE_K |
157 |
|
- GM_NON_UNITY_DIAGONAL |
158 |
|
- GM_EXTRA_DIAGONAL |
159 |
|
- GM_BOLUS_ADVEC ) |
160 |
|
* However, wrong gradient check problem remains unsolved. |
161 |
|
* New CPP options have been introduced for different |
162 |
|
tapering schemes |
163 |
|
o external_fields_load now part of differentiation list |
164 |
|
o pressure needs multiple storing; |
165 |
|
would be nice to have store_pressure at beginning or |
166 |
|
end of forward_step, e.g. by having phiHyd global (5-dim.) |
167 |
|
(NB: pressure is needed for certain cases in find_rho, |
168 |
|
which is also invoked through convective_adjustment). |
169 |
|
o recomputations in find_rho for cases |
170 |
|
'JMD95'/'UNESCO' or 'MDJWF' are OK. |
171 |
|
o #define ATMOSPHERIC_LOADING should be differentiable |
172 |
|
o ini_forcing shifted to begining of initialise_varia |
173 |
|
o Incorporating QNVS line search routines into MITgcm |
174 |
|
(this is separate code, and therefore not under pkg) |
175 |
|
lsopt/, optim/ |
176 |
|
o Updated verification/carbon/ |
177 |
|
- #define GM_VISBECK_VARIABLE_K |
178 |
|
- GM_taper_scheme = 'ldd97' |
179 |
|
|
180 |
checkpoint46n_post |
checkpoint46n_post |
181 |
Merging from release1_p8: |
Merging from release1_p8: |
182 |
o verification/natl_box: |
o verification/natl_box: |