3 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
4 |
============================== |
============================== |
5 |
|
|
6 |
|
o incorporating changes to get TAMC/TAF working in genmake2 |
7 |
|
- added new TAMC/TAF targets to the Makefile |
8 |
|
- various cleanups related to package handling |
9 |
|
|
10 |
|
checkpoint51l_post |
11 |
|
o changes to pkg/cal and pkg/exf that allow and speed-up long integrations |
12 |
|
- faster algorithm for cal_addtime.F |
13 |
|
- higher precision arithmetic, where needed |
14 |
|
- pkg/exf *startdate variables changed from calendar date to |
15 |
|
model integration time |
16 |
|
|
17 |
|
checkpoint51l_pre |
18 |
|
o add salt flux to forcing fields (used when salty sea-ice forms or melts) |
19 |
|
o enable to apply restoring on S & T only in a latitude band. |
20 |
|
o update output (external_forcing_surf modif. change truncation errors) |
21 |
|
global_ocean.cs32x15, global_ocean_pressure, lab_sea, natl_box |
22 |
|
|
23 |
|
checkpoint51k_post |
24 |
|
o change default to: useNHMTerms=.FALSE. |
25 |
|
- new output.txt for atmos. exp (NHMTerms were wrong but now turned off): |
26 |
|
adjustment.128x64x1, aim.5l_Eq_Ch, aim.5l_LatLon, hs94.128x64x5, hs94.1x64x5 |
27 |
|
- new output.txt for idealized oceanic exp (NHMTerms are now turned off): |
28 |
|
exp1, ideal_2D_oce |
29 |
|
- change data file of oceanic exp (leave the results unchanged for now) |
30 |
|
exp2, global_ocean[90x40x15,pressure,with_exf], lab_sea, natl_box |
31 |
|
o set gravitySign automatically according to the vertical coordinate used |
32 |
|
(and change sign: back to "logical" value: +1 when R increases downward) |
33 |
|
o added an include barrier to the default CPP_OPTIONS.h and |
34 |
|
removed the #include "PACKAGES_CONFIG.h" from it |
35 |
|
o fix a bug in mom_fluxform (from chk51j_post) |
36 |
|
|
37 |
|
checkpoint51j_post |
38 |
|
o Added some AD-related initialisations in mom_vecinv/ mom_fluxform/ |
39 |
|
o adjusted some flow directives (dynamics, thermodynamics) |
40 |
|
o added cubed-sphere case to test cost function |
41 |
|
o heckpoint_lev?_directives.h now include |
42 |
|
pkg-specific directives, whenever those are enabled. |
43 |
|
o Moving pkg-specific flow directives to corresponding pkg |
44 |
|
Moving pkg-specific differentiation list to corresponding pkg |
45 |
|
Moving pkg-specific checkpoint-lev storing to corresponding pkg |
46 |
|
---> conventions are |
47 |
|
- pkg_ad.flow |
48 |
|
- pkg_ad_diff.list |
49 |
|
- pkg_ad_check_lev?_dir.h |
50 |
|
|
51 |
|
checkpoint51i_post |
52 |
|
o merge of the branch-genmake2 |
53 |
|
- knowingly breaks the older genmake |
54 |
|
- automatic generation of all ALLOW_${PKG_NAME} defines |
55 |
|
- new testreport supersedes older testscript |
56 |
|
o large number of package inclusion and header cleanups |
57 |
|
checkpoint51i_pre |
58 |
|
|
59 |
|
o new output for adjoin of global ocean vector invariant |
60 |
|
o fix for checkpoint of thermodynamic seaice and bulk_force |
61 |
|
o Steph gchem various cleanups |
62 |
|
|
63 |
|
checkpoint 51h_pre |
64 |
|
o fixed some single quotes |
65 |
|
o modify gchem pkg to be more versatile |
66 |
|
o modifications to clean up dic pkg |
67 |
|
o add pkg/cfc |
68 |
|
|
69 |
|
checkpoint51g_post |
70 |
|
o provide links for ARPACK and ADM*TLM |
71 |
|
o start removing verification/vero/ |
72 |
|
o update AD-related stuff |
73 |
|
(in particular some IF-statements in model/src/ routines) |
74 |
|
o adding to verification/global_ocean.90x40x15/ adjoint of |
75 |
|
vector invariant setup |
76 |
|
|
77 |
|
checkpoint51f_post |
78 |
|
o update advect_xz/output.txt (left from chekpoint50f_pre, ini_vel modif) |
79 |
|
o Mods and bug fixes to pkg/cal, pkg/exf, etc., needed for computation |
80 |
|
of tracer Green's fucntions for ocean inversion project. |
81 |
|
|
82 |
|
checkpoint51f_pre |
83 |
|
o Added on-the-fly spatial interpolation capability |
84 |
|
"USE_EXF_INTERPOLATION" to pkg/exf. |
85 |
|
- This is a temporary Cartesian-grid hack until |
86 |
|
the super-duper ESMF coupler becomes available. |
87 |
|
- See verification/global_with_exf/README for usage example. |
88 |
|
- Removed obsolete EXFwindOnBgrid and SEAICEwindOnCgrid |
89 |
|
flags and modified pkg/seaice accordingly. |
90 |
|
o Bug fix to pkg/ptracers, pkg/generic_advdiff/gad_calc_rhs.F, |
91 |
|
and pkg/kpp/kpp_transport_ptr.F for dealing with tracer |
92 |
|
non-local transport term. |
93 |
|
|
94 |
|
checkpoint51e_post |
95 |
|
o pkg/mom_vecinv: |
96 |
|
- changes related to hFac: use S/R from aim.5l_cs/code (coriolis & vort.advec) |
97 |
|
- add time-ave diagnostic of vorticity advection. |
98 |
|
o add diagnostic (instantaneous) of Shapiro Filter effect for T,S & UV. |
99 |
|
o atmospheric geopotential: - include water vapor effect; |
100 |
|
- use p^kappa in p* scaling. |
101 |
|
o land pkg: read initial state from files. |
102 |
|
o AIM : add stratospheric drag in the upper level |
103 |
|
|
104 |
|
checkpoint51d_post |
105 |
|
o change function ILNBLNK (=> last non-blank char.) to return zero |
106 |
|
(instead of the string length) when the string. char is empty |
107 |
|
o fix bug in dxV,dyU definition of cubed-sphere grid (ini_curvilinear_grid) |
108 |
|
o modif's to restore TAMC compatibility |
109 |
|
- avoid "_d" constructs in PARAMETER declarations |
110 |
|
- avoid interrupting lines by comment lines |
111 |
|
o Use cluster local disks for purely local I/O |
112 |
|
vs. globally visible disks needed for ctrl stuff |
113 |
|
(merged from ecco-branch) |
114 |
|
- new active_file_loc... and mdsio_..._loc routines |
115 |
|
- affected packages: autodiff, ctrl, grdchk, mdsio |
116 |
|
|
117 |
|
checkpoint51c_post |
118 |
|
o introducing integer flag debugLevel |
119 |
|
- debLevZero: suppress all output |
120 |
|
- debLevA: standard |
121 |
|
- debLevB: equivalent to current debugMode=.TRUE. |
122 |
|
o introducing pathname variable mdsioLocalDir for mdsio |
123 |
|
that can be specified at runtime |
124 |
|
(relevant for cluster I/O to local disk) |
125 |
|
` |
126 |
|
>>>checkpoint51b_post |
127 |
|
adjoint of ptracers |
128 |
|
o disentangled ALLOW_PTRACERS using new ALLOW_GCHEM |
129 |
|
o extended state in checkpoint_lev?, etc. |
130 |
|
o alternative ctrl/cost with tr1 or ptracer |
131 |
|
o new comlev1_bibj_ptracers |
132 |
|
o new ptracers.flow |
133 |
|
o made ptracers_forcing same as external_forcing_s |
134 |
|
o added verification for adjoint with ptracers |
135 |
|
- adm. grdchk fail so far |
136 |
|
- tlm. grdchk OK |
137 |
|
- checked tapelev/comlev recomputations: OK |
138 |
|
<<<checkpoint51b_pre |
139 |
|
|
140 |
|
checkpoint51a_post |
141 |
|
o added code for biogeochemistry package |
142 |
|
- new pkgs: dic, gchem |
143 |
|
- new subroutine in ptracers |
144 |
|
o modify code for biogeo packages |
145 |
|
- in model/src and pkg/ptracers |
146 |
|
|
147 |
|
checkpoint51 |
148 |
|
o merged improved packages from ecco-branch (ecco_c51_e34) to MAIN |
149 |
|
- autodiff, ctrl, exf, grdchk, seaice |
150 |
|
kept separate packages |
151 |
|
- cost, ecco, obcs |
152 |
|
o updating verification/internal_wave, verification/lab_sea |
153 |
|
(all verif's pass on faulks.lcs.mit.edu RH7.3, g77) |
154 |
|
|
155 |
|
checkpoint50i_post |
156 |
|
o Preparing next differentiable checkpoint and sync |
157 |
|
of MAIN vs. ecco-branch |
158 |
|
- 1st step: this one is differentiable |
159 |
|
- 2nd step: next(! not this) checkpoint will have updated packages |
160 |
|
from ecco-branch) |
161 |
|
|
162 |
|
checkpoint50h_post |
163 |
|
o new package "land": provide surface BC for AIM. |
164 |
|
o pkg/ptracers: apply exchange to all tracers after reading a pickup. |
165 |
|
o AIM: use true log-P extrapolation for near surface temp (= as in |
166 |
|
F.Molteni paper) ; update aim.5l_cs output. |
167 |
|
o New variable in PARM03 for more flexible pickups |
168 |
|
o somewhat cleaned package initialisation sequence for |
169 |
|
ctrl/ cost/ ecco/ in packages_... routines |
170 |
|
|
171 |
|
checkpoint50g_post |
172 |
|
o merged with release1_p17 (pkg/seaice and verification/lab_sea) |
173 |
|
- added SEAICE_MULTILEVEL for 8-category sea-ice thermodynamics |
174 |
|
- LSR sea-ice dynamic solver moved to SouthWest B-grid location and |
175 |
|
made the default because of faster convergence than ADI |
176 |
|
|
177 |
|
checkpoint50f_post |
178 |
|
o exf totally restructured |
179 |
|
- replaced all exf_set_ routines by one generic exf_set_gen.F |
180 |
|
and made sure it's still differentiable |
181 |
|
- split exf_getffields.F into new exf_bulkformulae.F |
182 |
|
- exf_inscal_/exf_outscal_ suffixes consistent with field names |
183 |
|
- updated adjoint/makefile and relevant verif. accordingly |
184 |
|
|
185 |
|
checkpoint50f_pre |
186 |
|
o AIM: near surface temp evaluated from Pot.Temp: fixed for Part Cell ; |
187 |
|
+ change Evap over land. (update aim.5l_cs output) |
188 |
|
o added missing .cvsignores and made uniform across expts |
189 |
|
o changed configuration of tidal_basin_2d expt |
190 |
|
o added pkg/debug functions for tracing code in fatal crashes |
191 |
|
o tidied pkg/monitor to use MONITOR_OPTIONS.h |
192 |
|
o new fn in pkg/monitor to stop model if solution grows too big |
193 |
|
o removed lots of single quotes from comments |
194 |
|
|
195 |
|
checkpoint50e_post |
196 |
|
o make KPP work with PTRACERS |
197 |
|
- fix gad_calc_rhs to call new routine kpp_transport_ptr, which is |
198 |
|
nearly a copy of kpp_transport_s |
199 |
|
- there is no analogue to SurfaceTendencyS, so I have to use |
200 |
|
gPtr(of the surface layer) instead |
201 |
|
o add a new platform SunFire+mpi (SunFire 15000) to genmake |
202 |
|
checkpoint50e_pre |
203 |
|
|
204 |
|
checkpoint50d_post |
205 |
|
o change kpp output from multiple-record state files to single-record state |
206 |
|
files analogous to write_state.F |
207 |
|
o reduce the output frequency of cg3d-related stuff to the monitor frequency, |
208 |
|
analogous to the cg2d-related output. |
209 |
|
o fix small problem with in ptracers_write_checkpoint.F: len(suff)=512, |
210 |
|
so that writing to internal file fn (with length 512) fails. |
211 |
|
checkpoint50d_pre |
212 |
|
|
213 |
|
checkpoint50c_post |
214 |
|
o fix problem in ini_procs.F when using MPI (was there since checkpoint48f) |
215 |
|
o add monitor diagnostic of vorticity (mon_vort3.F) and potential energy. |
216 |
|
o add time-average diagnostic of U*V and vertical diffusive flux for theta. |
217 |
|
o bug fixed in convective adjustment diagnostic (multi-tile set-up). |
218 |
|
o 4th Order Adv scheme: return to previous (c40pre7) masking near boundaries. |
219 |
|
|
220 |
|
checkpoint50c_pre |
221 |
|
Merging from release1_p13: |
222 |
|
o bug fix for pkg/seaice dynamic solver |
223 |
|
o Added SEAICE_initialHEFF to pkg/seaice |
224 |
|
|
225 |
|
checkpoint50b_post |
226 |
|
o store u*,v* in gU,V instead of in guNm1,gvNm1 |
227 |
|
o in order to put the momForcing out of the Adams-Bashforth: |
228 |
|
move forcing & CD-scheme calls from mom_fluxform & mom_vecinv |
229 |
|
to timestep.F |
230 |
|
o new flag "useCDscheme" (default=F); replace guCD,gvCD by local arrays |
231 |
|
o re-generate output for highly sensitive test-exp: |
232 |
|
global_ocean_pressure, ideal_2D_oce, lab_sea, natl_box |
233 |
|
|
234 |
|
checkpoint50b_pre |
235 |
|
o restore the use of momAdvection & useCoriolis flags (mom_vecinv) |
236 |
|
o corrected CPP option mismatch in exf_Bulk... functions for case |
237 |
|
#define ALLOW_ATM_WIND && #undef ALLOW_ATM_TEMP |
238 |
|
|
239 |
|
checkpoint50a_post |
240 |
|
o Bug fix for merging between c50 and KPP. |
241 |
|
ikey was passed from thermodynamics to kpp_calc via |
242 |
|
common block rather than being recomputed in kpp_calc, |
243 |
|
in contradiction with new key itdkey. |
244 |
|
New key ikppkey created, and tamc.h headers updated. |
245 |
|
|
246 |
|
checkpoint50 |
247 |
|
- 2nd step of merge with ecco-branch: |
248 |
|
merge changes in ecco-branch back into main trunk |
249 |
|
- bug fix in find_alpha |
250 |
|
- more storing in gmredi_calc_tensor.F, kpp_routines.F |
251 |
|
- avoid iterative loop in ini_pressure for ALLOW_AUTODIFF_TAMC |
252 |
|
- merged some sliced AD support routines |
253 |
|
|
254 |
|
checkpoint49 |
255 |
|
merged from ecco-branch: |
256 |
|
o exf: |
257 |
|
- Enable initialisation of forcing fields to constant |
258 |
|
(runtime) values. |
259 |
|
- in exf_getffields.F |
260 |
|
Reduce i-/j-loop to interior domain, discarding overlaps. |
261 |
|
That also fixes wrong TAF-key computations for key_1, key_2 |
262 |
|
with bulf formulae. |
263 |
|
- exf_init.F modify #ifdef for exf_init_evap |
264 |
|
- exf_getffieldrec.F, ctrl_getrec.F |
265 |
|
The following INT-usages are not safe: |
266 |
|
fldsecs = int(fldsecs/fldperiod)*fldperiod |
267 |
|
fldcount = int(fldsecs/fldperiod) + 1 |
268 |
|
and were modified. |
269 |
|
o autodiff: checkpoint_lev... |
270 |
|
- updated storing for bulk formulae (some arrays were missing) |
271 |
|
- added state for seaice package |
272 |
|
- synchronized with main branch |
273 |
|
o DIVA: |
274 |
|
Modif.s in |
275 |
|
- the_main_loop.F |
276 |
|
- the_model_main.F |
277 |
|
and added cost_final_restore.F cost_final_store.F |
278 |
|
o With these changes, it is possible to fully separate |
279 |
|
the first forward run to compute the cost function (mdthe_main_loop) |
280 |
|
and adjoint runs over individual DIVA intervals (adthe_main_loop) |
281 |
|
into separate model executions. |
282 |
|
This enables better tuning of DIVA to given queue sizes. |
283 |
|
o Loss of cost function value in consecutive DIVA runs is |
284 |
|
fixed by storing to 'costfinal' file. |
285 |
|
o Appropriate call of ctrl_unpack/_pack,grdchk_main at |
286 |
|
beginning/end of adjoint integration is accounted for. |
287 |
|
The current changes apply to TAF-1.4.26. |
288 |
|
TAF has one remaining bug, requiring one manual change |
289 |
|
(see MITgcm manual). |
290 |
|
o cal: synchronized between MAIN and ecco-branch |
291 |
|
|
292 |
|
checkpoint48i_post |
293 |
|
o Changes to restore differentiability of code w.r.t. previous tag |
294 |
|
(mostly adding new routines to make list and replacing |
295 |
|
pressure by totPhiHyd). |
296 |
|
o Updated and merged grdchk package |
297 |
|
- has both ADM and TLM checks |
298 |
|
- works for single- and multi-proc. |
299 |
|
- output cleaned |
300 |
|
- worked successfully for parallel DIVA |
301 |
|
o genmake: |
302 |
|
- Adding -platform=SP4 |
303 |
|
- Modif. for DEC Alpha (many don't have makedepend; use mkdep instead). |
304 |
|
o Modif. for DIVA: |
305 |
|
ctrl_unpack should not be called in intermediate DIVA interval. |
306 |
|
(similar thing would apply for active files such as |
307 |
|
tbar, sbar, psbar, which should not be reinitialised; |
308 |
|
only relevant for ecco-branch) |
309 |
|
|
310 |
checkpoint48h_post |
checkpoint48h_post |
311 |
o improve "ini_p_ground" when using finite difference form for calc_phi_hyd |
o improve "ini_p_ground" when using finite difference form for calc_phi_hyd |
312 |
o Added net flux to downward flux conversion to pkg/exf/exf_getffields.F |
o Added net flux to downward flux conversion to pkg/exf/exf_getffields.F |