3 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
4 |
============================== |
============================== |
5 |
|
|
6 |
|
o change global_ocean.cs32x15 because of Pb with GM on CS-grid. |
7 |
|
o changes to permit hi-res, cubed-sphere, configuration |
8 |
|
- added useSingleCpuIO capability to mdsio_readfield.F and exf_interp.F |
9 |
|
- added "#undef ALLOW_USE_MPI" support to eesupp/src/scatter_2d.F |
10 |
|
- added pkg/exf/exf_set_uv.F for on-the-fly interpolation and rotation |
11 |
|
of surface winds for the cube |
12 |
|
(A verification experiment, based on global_ocean.cs32x15, but with |
13 |
|
pkg/seaice turned on and with on-the-fly interpolation from the NCEP |
14 |
|
Gaussian grid is described in MITgcm_contrib/high_res_cube/README_ice, |
15 |
|
complete with example surface forcing files and matlab scripts to look |
16 |
|
at the output and compare it with that of global_ocean.cs32x15.) |
17 |
|
o packages pickup: $PKG_write_pickup called from packages_write_pickup ; |
18 |
|
but call to $PKG_read_pickup remains in $PKG_ini_vars: change CD_code |
19 |
|
to conform to the common rule. |
20 |
|
|
21 |
|
checkpoint52d_pre |
22 |
|
o surface flux time-av diagnostic modified for NLFS & RealFreshWater |
23 |
|
o add call to AIM pendant S/R in external_forcing & external_fields_load.F |
24 |
|
|
25 |
|
checkpoint52c_post |
26 |
|
o Fixing OPTIONS files for use of exf: |
27 |
|
- ALLOW_ECCO not needed |
28 |
|
- ECCO_CPPOPTIONS.h should not be included from |
29 |
|
EXF_OPTIONS.h/CAL_OPTIONS.h, but from CPP_OPTIONS.h |
30 |
|
as has been the case in the past |
31 |
|
- All cal/exf related options are currently set in |
32 |
|
single header file. Should remain so until we |
33 |
|
agree on a consistent new policy. |
34 |
|
o Removed all CPP options for cal package. |
35 |
|
cal has no more CPP options. |
36 |
|
o modified STOREs in GAD_ADVECTION |
37 |
|
o corrected key comp. for passkey |
38 |
|
|
39 |
|
checkpoint52b_post |
40 |
|
o new version: "thsice" of thermodynamic sea-ice pkg: |
41 |
|
- only 1 call from "forward_step"; conserve energy, fresh-water & salt. |
42 |
|
- modify fluxes but do not change theta directly; |
43 |
|
- light cleaning of bulk_force pkg (still needed with thsice). |
44 |
|
- add forcing fields + data files in global_ocean.cs32x15/inp_thsice |
45 |
|
+ output_thsice.txt to test thsice & bulk_force pkgs. |
46 |
|
o put back grid & coriolis monitor output. |
47 |
|
checkpoint52b_pre |
48 |
|
|
49 |
|
o call "do_the_model_io" at the beginning (=> dump initial state) |
50 |
|
o use rhoConstFresh to define convertEmP2rUnit (was done only in ocean-P) |
51 |
|
o sort out interplay between tamc.h and PTRACERS_OPTIONS.h |
52 |
|
(need info on NUMBER_OF_PTRACERS) |
53 |
|
o missing PTRACERS headers in some routines |
54 |
|
o default tamc.h in pkg/autodiff supposed to crash |
55 |
|
to ensure proper customization |
56 |
|
|
57 |
|
checkpoint52a_post |
58 |
|
o modifications to make FREEZE (_SURFACE) flux visible to pkg/kpp |
59 |
|
- moved surfaceTendencyTice from pkg/seaice to main code |
60 |
|
- subroutine FREEZE_SURFACE now limits only surface temperature |
61 |
|
and is called in FORWARD_STEP (if allowFreezing=.TRUE.) |
62 |
|
=> new output.txt for global_ocean.cs32x15 & global_with_exf |
63 |
|
- keep the old version of FREEZE for backward compatibility; |
64 |
|
global_ocean.90x40x15 left unchanged using useOldFreezing=.TRUE. |
65 |
|
- corresponding modifs for TAF directives |
66 |
|
o added surface flux output variables to TIMEAVE_STATVARS |
67 |
|
o time-averaged output for pkg/ptracers |
68 |
|
o fix problem with ECCO-related divided adjoint |
69 |
|
snapshot file was missing some ad-variables, |
70 |
|
but no warning issued by TAF |
71 |
|
|
72 |
|
checkpoint52a_pre |
73 |
|
o some bug fixes for #undef REAL4_IS_SLOW |
74 |
|
|
75 |
|
checkpoint52 = ecco_c52_e35 |
76 |
|
o after testing adjoints of checkpoint51u_post |
77 |
|
|
78 |
|
checkpoint51u_post |
79 |
|
o this is the post-merge tag of the ecco-branch |
80 |
|
(needs another clean checkout and re-testing) |
81 |
|
o various natl_box_adjoint configs and results |
82 |
|
but no input fields |
83 |
|
o obcs adjoint not yet merged! |
84 |
|
o adding ECCO global 1x1 and 2x2 degree configs |
85 |
|
o updated/extended some store directives for seaice |
86 |
|
o pkg/ecco now containes ecco-specific part of cost function |
87 |
|
o top level routines the_main_loop, forward_step |
88 |
|
supersede those in model/src/ |
89 |
|
previous input data.cost now in data.ecco |
90 |
|
(new namelist ecco_cost_nml) |
91 |
|
o in ctrl/ cleaned some cross-dependencies and updated CPP options |
92 |
|
|
93 |
|
checkpoint51t_post |
94 |
|
o this is the pre-merge tag of the ecco-branch |
95 |
|
|
96 |
|
checkpoint51s_post |
97 |
|
o cleanup: removed various obsolete #define-s |
98 |
|
o pkg/grdchk |
99 |
|
- renamed 'ph-grd 3' to 'grad-res' |
100 |
|
- merged from ecco-branch |
101 |
|
(remaining bug fixes for obcs gradient checks) |
102 |
|
- additional high-precision output for testreport |
103 |
|
(grep for 'precision_grdchk_result') |
104 |
|
|
105 |
|
checkpoint51r_post |
106 |
|
o cleaned up or removed CPP_OPTIONS.h files |
107 |
|
o updated results for some of the tests to reflect the new |
108 |
|
optfiles (-ieee: -O0 instead of -O3) |
109 |
|
|
110 |
|
checkpoint51q_post |
111 |
|
o cd_code: |
112 |
|
- moved cd_scheme.F -> cd_code_scheme.F |
113 |
|
- separate read_checkpoint from cd_code_ini_vars.F |
114 |
|
- separated cd_code part from write_checkpoint |
115 |
|
- updated AD_SOURCE, generated .flow |
116 |
|
- added CD_CODE_VARS.h to the_main_loop |
117 |
|
- Moving cd_code specific store dir.s to pkg/cd_code |
118 |
|
o ctrl package totally restructured |
119 |
|
- pack/unpack now optional and decoupled from |
120 |
|
xx_/adxx_ I/O |
121 |
|
- ctrl_pack/unpack cleaned |
122 |
|
(new routines ctrl_init_ctrlvar.F, pkg/ctrl/ctrl_init_wet.F) |
123 |
|
- confined inclusion of AD_CONFIG.h to where necessary. |
124 |
|
o genmake2: changed AD dependencies from SRCFILES to AD_FILES |
125 |
|
|
126 |
|
checkpoint51p_post |
127 |
|
o create initial cd_code package |
128 |
|
- WARNING: the initialization of fields within packages |
129 |
|
is broken because INI_FIELDS (where READ_CHECKPOINT is |
130 |
|
called) is called *before* PACKAGES_INIT_VARIABLES. |
131 |
|
This seems to be backwards since |
132 |
|
1) its important to initially zero the entire field including |
133 |
|
the "halo" region |
134 |
|
2) zeroing needs to be done *before* reading pickup info |
135 |
|
and the pickup call might or might not happen in any |
136 |
|
particular run |
137 |
|
- this needs an overhaul for R2 |
138 |
|
o cleanup verification/*/code/GMREDI_OPTIONS.h |
139 |
|
o cleanup testreport |
140 |
|
|
141 |
|
checkpoint51o_post |
142 |
|
o bringing AD-related verif's up to date with genmake2 |
143 |
|
- carbon/ |
144 |
|
- global_ocean.90x40x15/ |
145 |
|
- hs.94.1x64x5/ |
146 |
|
- (front_relax/ ) |
147 |
|
o (more) consistent directory structure for AD-related verif.'s |
148 |
|
o cleaning ALLOW_GRADIENT_CHECK -> ALLOW_GRDCHK |
149 |
|
o cleaning some ALLOW_TANGENTLINEAR_RUN -> ALLOW_AUTODIFF |
150 |
|
o bug fix in find_alpha.F for MDJWF: |
151 |
|
- modif. to alpha = 1/D*( dN/dT - rho*dD/Dt) to account for |
152 |
|
change rho -> rho-rhoConst |
153 |
|
- replace call find_rho to find_rhonum |
154 |
|
o initialisation of rFlx extended to full array (required by TAF) |
155 |
|
and shifted to thermodynamics |
156 |
|
o removed PTRACERS.h in ptracers routine |
157 |
|
o added surfacetendencyPtr to S/R parameter list pracers_forcing |
158 |
|
o Correcting initialisations for fVerU, fVerV. |
159 |
|
o helping TAF to recognize dependency on kUp for |
160 |
|
fVerU, fVerV |
161 |
|
o Replacing delZ by delR in pkg/ctrl/ |
162 |
|
checkpoint51o_pre |
163 |
|
|
164 |
|
o undid all of cp51 pending some cleanups and discussion |
165 |
|
o added the [#include "AD_CONFIG.h"] statement to all files that need |
166 |
|
it for adjoint/tl #defines |
167 |
|
o re-worked the build logic in genmake2 to support AD_CONFIG.h |
168 |
|
o removed tools/genmake since it no longer works |
169 |
|
|
170 |
|
checkpoint51n_post |
171 |
|
o modifications to make FREEZE flux visible to pkg/kpp |
172 |
|
- moved surfaceTendencyTice from pkg/seaice to main code |
173 |
|
- FREEZE & EXTERNAL_FORCING_SURF moved to FORWARD_STEP |
174 |
|
- subroutine FREEZE now limits only surface temperature |
175 |
|
(this means new output.txt for global_ocean.90x40x15, |
176 |
|
global_ocean.cs32x15, and global_with_exf) |
177 |
|
o added surface flux output variables to TIMEAVE_STATVARS |
178 |
|
|
179 |
|
checkpoint51n_pre |
180 |
|
o added "#undef ALLOW_USE_MPI" support to eesupp/src/gather_2d.F |
181 |
|
o added [#include "AD_CONFIG.h"] to all files that use the following |
182 |
|
defines: |
183 |
|
- ALLOW_ADJOINT_RUN |
184 |
|
- ALLOW_TANGENTLINEAR_RUN |
185 |
|
- ALLOW_ECCO_OPTIMIZATION |
186 |
|
|
187 |
|
checkpoint51m_post |
188 |
|
o incorporating changes to get TAMC/TAF working in genmake2 |
189 |
|
- added new TAMC/TAF targets to the Makefile |
190 |
|
- various cleanups related to package handling |
191 |
|
- these changes should have no impact on the forward model |
192 |
|
|
193 |
|
checkpoint51l_post |
194 |
|
o changes to pkg/cal and pkg/exf that allow and speed-up long integrations |
195 |
|
- faster algorithm for cal_addtime.F |
196 |
|
- higher precision arithmetic, where needed |
197 |
|
- pkg/exf *startdate variables changed from calendar date to |
198 |
|
model integration time |
199 |
|
|
200 |
|
checkpoint51l_pre |
201 |
|
o add salt flux to forcing fields (used when salty sea-ice forms or melts) |
202 |
|
o enable to apply restoring on S & T only in a latitude band. |
203 |
|
o update output (external_forcing_surf modif. change truncation errors) |
204 |
|
global_ocean.cs32x15, global_ocean_pressure, lab_sea, natl_box |
205 |
|
|
206 |
|
checkpoint51k_post |
207 |
|
o change default to: useNHMTerms=.FALSE. |
208 |
|
- new output.txt for atmos. exp (NHMTerms were wrong but now turned off): |
209 |
|
adjustment.128x64x1, aim.5l_Eq_Ch, aim.5l_LatLon, hs94.128x64x5, hs94.1x64x5 |
210 |
|
- new output.txt for idealized oceanic exp (NHMTerms are now turned off): |
211 |
|
exp1, ideal_2D_oce |
212 |
|
- change data file of oceanic exp (leave the results unchanged for now) |
213 |
|
exp2, global_ocean[90x40x15,pressure,with_exf], lab_sea, natl_box |
214 |
|
o set gravitySign automatically according to the vertical coordinate used |
215 |
|
(and change sign: back to "logical" value: +1 when R increases downward) |
216 |
|
o added an include barrier to the default CPP_OPTIONS.h and |
217 |
|
removed the #include "PACKAGES_CONFIG.h" from it |
218 |
|
o fix a bug in mom_fluxform (from chk51j_post) |
219 |
|
|
220 |
|
checkpoint51j_post |
221 |
|
o Added some AD-related initialisations in mom_vecinv/ mom_fluxform/ |
222 |
|
o adjusted some flow directives (dynamics, thermodynamics) |
223 |
|
o added cubed-sphere case to test cost function |
224 |
|
o heckpoint_lev?_directives.h now include |
225 |
|
pkg-specific directives, whenever those are enabled. |
226 |
|
o Moving pkg-specific flow directives to corresponding pkg |
227 |
|
Moving pkg-specific differentiation list to corresponding pkg |
228 |
|
Moving pkg-specific checkpoint-lev storing to corresponding pkg |
229 |
|
---> conventions are |
230 |
|
- pkg_ad.flow |
231 |
|
- pkg_ad_diff.list |
232 |
|
- pkg_ad_check_lev?_dir.h |
233 |
|
|
234 |
checkpoint51i_post |
checkpoint51i_post |
235 |
o merge of the branch-genmake2 |
o merge of the branch-genmake2 |
236 |
- knowingly breaks the older genmake |
- knowingly breaks the older genmake |