3 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
4 |
============================== |
============================== |
5 |
|
|
6 |
|
o Added cost_ice for sea-ice volume sensitivity experiment. |
7 |
|
o ecco_the_model_main.F: SEAICE_WRITE_PICKUP must precede WRITE_CHECKPOINT |
8 |
|
|
9 |
|
icebear3 |
10 |
|
o Removed spurious iikey and idkey |
11 |
|
o pkg/exf |
12 |
|
- bug fix in treatment of downward radiation |
13 |
|
o pkg/seaice |
14 |
|
- added initial sea-ice thickness parameter SEAICE_initialHEFF |
15 |
|
o updated verification/natl_box_adjoint |
16 |
|
|
17 |
|
icebear2 |
18 |
|
o Added missing /* */ to CPP comments in pkg/seaice, pkg/exf, |
19 |
|
kpp_transport_t.F,and ecco_the_main_loop.F |
20 |
|
o adjoint-friendly modifications to pkg/seaice |
21 |
|
o added verification/lab_sea as per release1_p12_pre |
22 |
|
|
23 |
|
ecco_c44_e26 |
24 |
|
o Merging pkg/exf and pkg/seaice from cvs tag ecco_ice2 |
25 |
|
o exf: Modifications to correctly handle evap field for adjoint |
26 |
|
o grdchk: |
27 |
|
Added code to perform gradient checks for bulk formulae/atmos. state |
28 |
|
(atemp, aqh, uwind, vwind). |
29 |
|
NOTE: |
30 |
|
This package is out of synch with c48 package. |
31 |
|
The latter also has the tangent linear gradient checks. |
32 |
|
Need to be merged. |
33 |
|
o verification/natl_box_adjoint: |
34 |
|
* added new setup: code_bulk/, input_bulk, results_bulk |
35 |
|
(Experiment 2, described in README) |
36 |
|
uses bulk formulae and atmos. state as control instead of |
37 |
|
air-sea fluxes. |
38 |
|
Performs gradient checks for first 4 elements of atemp control. |
39 |
|
* changed standard verification to perform |
40 |
|
gradient checks on first 4 elements on hflux control. |
41 |
|
* gmredi switched off (will come after merge with c48). |
42 |
|
|
43 |
|
ecco_c44_e25 |
44 |
|
o Merging some GM changes from main trunk (c47) to ecco-branch. |
45 |
|
o Updating verif. natl_box_adjoint with GMRedi. |
46 |
|
o Adding setup with explicit differentiaton of pressure solver. |
47 |
|
o Bug fix: missing OBCS_OPTIONS.h for sponge layer code. |
48 |
|
o (Re-?) Commiting old bug fixes in kpp to ecco-branch |
49 |
|
(kpp_calc.F, kpp_routines.F) |
50 |
|
o exf: |
51 |
|
updated external forcing package |
52 |
|
- cleaned old exfa stuff |
53 |
|
- bug fixes |
54 |
|
(missing OBCS_OPTIONS.h in two routines) |
55 |
|
- enable easy to use "no forcing". |
56 |
|
- added exf I/O for atmospheric loading |
57 |
|
- added exf I/O for runoff data |
58 |
|
- transfered scaling between exf <-> MITgcm to exf namelist |
59 |
|
|
60 |
|
ecco_c44_e24 |
61 |
|
o modifications to gradient check package (Martin Losch) |
62 |
|
- enable centered differences |
63 |
|
- modified format of standard output |
64 |
|
o bug fix for nwettile contents |
65 |
|
so far assumed global sum in first entry (1,1,k), |
66 |
|
needed for global control vector |
67 |
|
-> introduced new variable nwet[c/w/s]global(nr) |
68 |
|
o shifts of calls to cal/exf package to be accessible |
69 |
|
in MITgcm stand-alone |
70 |
|
o shifted packages_init_variables before convective_adjustment_ini |
71 |
|
(corresponding update of TAF tape initialisation). |
72 |
|
o removed auxiliary field in impldiff |
73 |
|
(caused numerical differences for GMRedi cases) |
74 |
|
o updated obcs control part (G. Gebbie) |
75 |
|
o removed some f90 comments ("!") |
76 |
|
o Bug fixes and updates in cost package |
77 |
|
* cost_final: correct multiplier mult_tauv, mult_sflux |
78 |
|
* cost_hyd: added call cost_ssh |
79 |
|
* cost_init_barfiles.F: replaced arrays by tmpfld's |
80 |
|
* cost_readtopexmean.F: ersmask -> tpmeanmask |
81 |
|
* cost_weights.F: corrected wtheta[2], wsalt[2] |
82 |
|
o cleaned exchanges in cg2d.F |
83 |
|
|
84 |
|
ecco_c44_e23 |
85 |
|
merged patches between release1_p1 and release1_p2: |
86 |
|
o fixed array boundary problem for adjoint |
87 |
|
(related to partial recomputation within subroutine |
88 |
|
related to dynamics -> calc_viscosity -> mom_... |
89 |
|
o fixed recomputation in thermodynamics |
90 |
|
o prepare (but not complete) recomputation problem in impldiff |
91 |
|
o fixed problem for adjoint of global_max |
92 |
|
o switched off sponge layer code in external_forcing |
93 |
|
Bug fix in ctrl_mask_set_yz (spotted by G. Gebbie) |
94 |
|
|
95 |
|
ecco_c44_e22 |
96 |
|
o Finally, the merges with the Scripps cost function terms: |
97 |
|
Various additions, modifications, quality checks. |
98 |
|
o Plus: Included switch ALLOW_AUTODIFF_MONITOR to handle |
99 |
|
monitoring of adjoint state. |
100 |
|
o Plus: removed comm_stats, ini_communication_patterns from diff. list |
101 |
|
(no more need to "keep" MPI headers for adjoint generation) |
102 |
|
o Plus: exf changes: |
103 |
|
o added run-off routine |
104 |
|
o additional storing needed for ustress,vstess |
105 |
|
because of state-dependent quality check |
106 |
|
o added missing initialisations |
107 |
|
o Plus: genmake with MAKEDEPEND variable was still missing in ecco-branch |
108 |
|
NB: - verification/natl_box_adjoint/ 240 steps run |
109 |
|
* looks ok under Linux (twain.lcs.mit.edu) |
110 |
|
* looks ok on MIT SUN's (fjord.mit.edu) |
111 |
|
* NaN in reverse mode on Scripps SUN's at iter=214 |
112 |
|
* NaN in reverse mode on IBM SP3 (BlueHorizon) at iter=204 |
113 |
|
- 1 year forward int. global 2x2 deg. produces similar cost functions |
114 |
|
* all cost terms of new code are slightly smaller (0.5 to 3%). |
115 |
|
|
116 |
|
ecco_c44_e21 |
117 |
|
o reversed order of ctrl/cost initialisation |
118 |
|
o gmredi_slope_psi.F: Inconsistent types in MIN/MAX expressions |
119 |
|
o modified to enable ctrl field swapping |
120 |
|
o bug fix for global sum of ncvarlength in ctrl_init |
121 |
|
o no more call to ctrl_unpack from ctrl_init |
122 |
|
o no more write of ctrl vector in ctrl_pack |
123 |
|
however, init. of control vector from ctrl_unpack |
124 |
|
o various bug fixes & consistent nobcs handling in |
125 |
|
ctrl_pack/unpack routines (by G. Gebbie) |
126 |
|
o optim package: Added obcs control part for lsopt I/O |
127 |
|
(by G. Gebbie) |
128 |
|
o missing call to obcs_diags |
129 |
|
o moved calls in packages_init_fixed to packages_readparms |
130 |
|
because some quantities are needed earlier. |
131 |
|
|
132 |
|
ecco_c44_e20 |
133 |
|
o before major merges from the Scripps setup |
134 |
|
here are two small updates: |
135 |
|
- field swapping for surface flux controls enabled |
136 |
|
(remains t.b.d. for atmos. state plus obcs) |
137 |
|
- bug fix in active_file.F for xz-/yz- routines |
138 |
|
- added missing xz-/yz- routines to active_file_ad.F |
139 |
|
|
140 |
|
ecco_c44_e19 |
141 |
|
o merged necessary patches: |
142 |
|
- AD-related changes for GMRedi |
143 |
|
- fixes i KPP (delZ -> drF) |
144 |
|
- hook to OBCS songe layer code in external_forcing |
145 |
|
- errorMessageUnit non-zero in eeboot.F |
146 |
|
- included GMREDI, EXACT_CONSERV to natl_box_adjoint verif. |
147 |
|
|
148 |
|
ecco_c44_e18 |
149 |
|
o sponge layer code for obcs was missing |
150 |
|
- added hook in external_forcing.F |
151 |
|
- added storing for TAF |
152 |
|
o new routine cost_init_barfiles to initialise |
153 |
|
adjoint fields adtbar, adsbar, adpsbar |
154 |
|
(needed in conjunction with corrected ladinit for active I/O) |
155 |
|
o cleaned up and corrected order in packages_readparms, ecco_readparms |
156 |
|
o corrected and simplified weights for obcs controls |
157 |
|
- weight fields simplified |
158 |
|
- bug fix in ctrl_pack |
159 |
|
- no more write of control vector in ctrl_pack |
160 |
|
o i/j index bug fixes in cost_ssh_mean |
161 |
|
|
162 |
|
ecco_c44_e17 |
163 |
|
o bug fixes in cost package |
164 |
|
* wrong identifier for active I/O of monthly mean psbar |
165 |
|
(cost_averagesfields, cost_ssh) xx_psbar_dummy |
166 |
|
* cleaned all identifier for active I/O of monthly fields |
167 |
|
* wrong ladinit in various cost routines |
168 |
|
o merge with interval (chkpt44c_post - chkpt44d_post) |
169 |
|
* added missing EXCLUDE_MONITOR flags |
170 |
|
* changed "e" to "_d" in gmredi_slope_limit, gmredi_slope_psi |
171 |
|
(incompatible typ in MIN/MAX expressions caused problems |
172 |
|
on IBM SP3) |
173 |
|
* in genmake added variable MAKEDEPEND |
174 |
|
plus resetting for case SunOS |
175 |
|
added timer_stats.c routine for IBM SP3 |
176 |
|
* removed variables in dynamics |
177 |
|
|
178 |
|
ecco_c44_e16 |
179 |
|
o new ECCO checkpoint, the first since ecco_c43_e15 |
180 |
|
which is branched from MITgcm. |
181 |
|
All changes were merged in between (see ecco-branch-mod? tags). |
182 |
|
o last changes: |
183 |
|
* assuming path for TAF and TAMC are known (e.g. in $path) |
184 |
|
* genmake: |
185 |
|
- added tools/xmakedepend to circumvene shortcoming of |
186 |
|
SUN's makedepend (size too small) |
187 |
|
- invoking ../tools/xmakedepend instead of makedepend |
188 |
|
- changed compiler flags for case SunOS |
189 |
|
|
190 |
|
ecco-branch-mod5 |
191 |
|
o mod's to reproduce open boundary run |
192 |
|
o enable field swapping for OB's and climatology |
193 |
|
o update store directives for OB control |
194 |
|
o added option ALLOW_OBCS_PRESCRIBE |
195 |
|
o call of getobcs now out of obcs_calc (param. list modified) |
196 |
|
|
197 |
|
ecco-branch-mod4 |
198 |
|
o this is the second part of changes necessary to reconcile |
199 |
|
the natl_box_adjoint verification (with and without grdchk). |
200 |
|
>>>!!! CHANGES BETWEEN THE TAGS ecco-branch-mod3 !!!<<< |
201 |
|
>>>!!! AND ecco-branch-mod4 !!!<<< |
202 |
|
>>>!!! SHOULD BE MERGED INTO THE MAIN TRUNK !!!<<< |
203 |
|
- genmake: removed $S64 overwrite for case SunOS |
204 |
|
- pkg/exf: update and corrections for field swapping and obcs |
205 |
|
- pkg/ecco: parameter lists for the_model_main, the_main_loop |
206 |
|
harmonized between ECCO and MITgcm |
207 |
|
- pkg/autodiff: added flow directives for obcs, mdsio_gl_slice |
208 |
|
updated checkpointing_lev... lists for obcs |
209 |
|
- model/src: minor changes in forward_step, plot_field |
210 |
|
- pkg/mdsio: added mdsio_gl_slice |
211 |
|
|
212 |
|
ecco-branch-mod3 |
213 |
|
o this is the first part of changes necessary to reconcile |
214 |
|
natl_box_adjoint verification (with and without grdchk). |
215 |
|
This set of changes is relevant for ECCO only. |
216 |
|
The next set/tag of changes (ecco-branch-mod4) |
217 |
|
should be merged into the main trunk. |
218 |
|
- update some options |
219 |
|
- parameter lists of ecco_the_model_main and ecco_the_main_loop |
220 |
|
differed from those of the MITgcm counterparts. |
221 |
|
|
222 |
|
ecco-branch-mod2 |
223 |
|
o update ecco-branch-mod1 to checkpoint44 |
224 |
|
o exclude from updates: |
225 |
|
- verification/ |
226 |
|
- adjoint/ |
227 |
|
- pkg/cost |
228 |
|
- pkg/ctrl |
229 |
|
- pkg/ecco |
230 |
|
- pkg/obcs |
231 |
|
- model/src/the_model_main.F |
232 |
|
- model/src/the_main_loop.F |
233 |
|
|
234 |
|
checkpoint44 |
235 |
|
o apply Shapiro filter on T&S before calc_PhiHyd (stagger time stepping) |
236 |
|
o fix bug in mom_vi_del2uv |
237 |
|
o added some stuff for AD |
238 |
|
|
239 |
|
ecco-branch-mod1 |
240 |
|
o based on c43 |
241 |
|
o replaced packages cost, ctrl, ecco, obcs |
242 |
|
o removed MITgcm verif.'s, added ECCO verif.'s |
243 |
|
o added lsopt, optim for line search optimization. |
244 |
|
|
245 |
|
checkpoint43 |
246 |
|
o added two new advection tests |
247 |
|
o added global 4x4 degree ocean test |
248 |
|
o added test of H&S94 with free surface on cube |
249 |
|
o some final mods for gradient check, adjoint,... |
250 |
|
|
251 |
|
checkpoint42 |
252 |
|
Fixed AD-related problems: |
253 |
|
o Store directives up-to-date with re-arranged Adams-Bashforth |
254 |
|
(mainly thermodynamics.F) |
255 |
|
o New store directives for multi-dim. advection schemes |
256 |
|
* new CPP flag ALLOW_MULTI_DIM_ADVECTION |
257 |
|
* new common block and key passkey |
258 |
|
(mainly gad_advection.F) |
259 |
|
o Modified store directives for split of dynamics/thermodynamics |
260 |
|
for the case ALLOW_KPP |
261 |
|
o Cleaned argument list for timestep_tracer.F |
262 |
|
|
263 |
|
checkpoint41 |
264 |
|
o re-formatted/added comments in prologues of key subroutines |
265 |
|
to work with protex |
266 |
|
|
267 |
|
checkpoint40 |
268 |
|
o moved calc_gs, calc_gt and calc_tr1 back to model/src |
269 |
|
o added FLT (floats) package from Arne along with demo. experiment |
270 |
|
o re-arranged Adams-Bashforth scheme for tracers to allow easier |
271 |
|
implementation of forward-in-time forcing terms |
272 |
|
o more mods for non-linear free-surface including new variable |
273 |
|
o modified multi-dim method to work on cube |
274 |
|
o bug fix in KPP (from DM) |
275 |
|
|
276 |
|
checkpoint40pre9 |
277 |
|
o added DST advection schemes |
278 |
|
o fix MPI bug on SGI's |
279 |
|
o modifications for non-linear free-surface |
280 |
|
o fixed conditionals for fixed flow/passive tracer only mode |
281 |
|
o added tension/strain form of stress tensor |
282 |
|
o changes to commenting of arguments for TAF/TAMC |
283 |
|
o added/updated many matlab scripts |
284 |
|
o added multi-dimensional advection method for tracers |
285 |
|
|
286 |
|
checkpoint40pre8 |
287 |
|
o new package rw/ |
288 |
|
Added method for dumping intermediate local arrays: |
289 |
|
mdsio_writetile - same as mdsio_writefield except works from inside |
290 |
|
bi,bj loop |
291 |
|
mdsio_writelocal - same as mdsio_writetile except works for local |
292 |
|
arrays |
293 |
|
write_local_r? - higher-level wrapper for mdsio_writelocal |
294 |
|
Controlled by diagFreq. Defaults to zero (ie. no dumps) |
295 |
|
Example given at end of mom_vecinv.F that dumps some local arrays. |
296 |
|
o replaced calc_mom_rhs.F by mom_fluxform.F/mom_vecinc.F to |
297 |
|
control flux-form versus vector invariant code. |
298 |
|
Switch is runtime flag vectorInvariantMomentum (defaults to false) |
299 |
|
or genmake -DISABLE ... |
300 |
|
o TAMC-related update to cope with mom_fluxform.F/mom_vecinc.F |
301 |
|
(additional storing is now needed for case vecinv; |
302 |
|
currently nothing implemented to switch this storing off |
303 |
|
in case fluxform is used since no CPP flag available). |
304 |
|
|
305 |
|
checkpoint40pre7 |
306 |
|
o bug fix in dynamics.F |
307 |
|
calc_viscosity called before k=1,Nr loop. |
308 |
|
o added some #ifdef's ALLOW_PASSIVE_TRACER |
309 |
|
o cleaned initialisations in ctrl_map_ini |
310 |
|
related to TAF |
311 |
|
|
312 |
|
checkpoint40pre6 |
313 |
|
o split dynamics into thermodynamics & dynamics |
314 |
|
o corresponding split of calc_diffusivity into |
315 |
|
calc_diffusivity, calc_viscosity |
316 |
|
(kpp_calc_diff, kpp_calc_visc). |
317 |
|
o Added diffkr, kapgm to control variables |
318 |
|
o bug fix in gmredi_slope_limit for ldd97 |
319 |
|
|
320 |
|
checkpoint40pre5 |
321 |
|
o proper initialization of Adams-Bashforth time stepping |
322 |
|
+ uses forward step for very first step of model (iter=0) |
323 |
|
o re-generated *all* output since almost all output is modified by this |
324 |
|
change (sorry but we've been putting this one off for about 2 years) |
325 |
|
+ natl_box, aim.5l_Equatorial_Channel and aim.5l_zon-ave were |
326 |
|
*not* regenerated since they are already out of date and I don't |
327 |
|
want to create the impression that they are suddenly working. |
328 |
|
o removed KAP stuff for DEC's |
329 |
|
o modified/added many .cvsignore files in */code and */input |
330 |
|
o added new expt "solid-body.cs-32x32x1" |
331 |
|
+ this uses an alternative form of ini_curvilinear_grid.F which |
332 |
|
is not necessarily any better than the current one... |
333 |
|
|
334 |
|
checkpoint40pre4 |
335 |
|
o added re-initialisations of utrans,vtrans,rtrans in |
336 |
|
calc_common_factors which are crucial for TAF. |
337 |
|
o extended iMin,jMin range for calc_common_factors, |
338 |
|
calc_diffusivity |
339 |
|
o added option ALLOW_PASSIVE_TRACER |
340 |
|
o a few minor updates and fixes |
341 |
|
|
342 |
|
checkpoint40pre3 |
343 |
|
o created calc_gtr1.F for gad package |
344 |
|
o split the_main_loop and re-introduced forward_step |
345 |
|
o some recomputations of nIter0 for TAMC |
346 |
|
|
347 |
|
checkpoint40pre2 |
348 |
|
o Updated to latest branch of AJA, JMC |
349 |
|
o Added gradient check package |
350 |
|
o Added routines for passive tracer handling |
351 |
|
o Added carbon sequestration sensitivity experiment |
352 |
|
|
353 |
|
checkpoint39 |
354 |
|
o Modifications and fixes to support TAMC |
355 |
|
(mainly missing or wrong directives) |
356 |
|
o Corrected common blocks when fooling TAMC to produce adjoint I/O |
357 |
|
in addummy_in_stepping |
358 |
|
o Updated verification/vero/ (bug fixes in code/, correct pickups in input/) |
359 |
|
o Enabled "sliced" (xz-,yz-) exchanges, needed for obcs. |
360 |
|
o Included following packages: |
361 |
|
cal: calendar |
362 |
|
exf: external forcing |
363 |
|
ecco: essential routines to control use of cal, exf |
364 |
|
(NB: this is not the full ECCO environment) |
365 |
|
o disabled some packages by default in genmake |
366 |
|
|
367 |
|
checkpoint38 |
368 |
|
o see notes_c37_adj.txt |
369 |
|
o corrected wrong layer index in external_forcing_t |
370 |
|
o replaced the_main_loop/forward_step by the_main_loop |
371 |
|
o removed forceoutput from write_state |
372 |
|
(replaced criteria for initial/final state) |
373 |
|
o added runtime flag for pre-C35 pickups: usePickupBeforeC35 |
374 |
|
|
375 |
|
checkpoint37 |
376 |
|
o mdsio.F routine split and packaged in pkg/mdsio |
377 |
|
(mdsio.F and mdsio_gl.F still exist in eesupp/src but are not |
378 |
|
used unless genmake -disable=mdsio) |
379 |
|
o allow a more precise surface-geopotential gradient for the atmosphere |
380 |
|
taking into account the change in space of Alpha (=1/rho) at the surface. |
381 |
|
o variable "etaN" (replace cg2d_x) is the surface R-anomaly [r unit] |
382 |
|
(ocean: sea surface height [m] / atmos: surface pressure anomaly [Pa]) |
383 |
|
o Change units: Phi_Hydrostatic (=phiHyd), Phi_Surface(=cg2d_x) and |
384 |
|
Phi_NonHydrostatic (=cg3d_x), all have now the dimension of a potential |
385 |
|
[m2/s2] (pressure/rhoConst in the ocean ; Geo-Potential in the atmosphere). |
386 |
|
o Time-Average diagnostics defined as a pseudo package "timeave". |
387 |
|
o pickup files : old pickup (before c35) did not work with new code (c35 |
388 |
|
and after) unless activating 2 commented lines in initialise-varia. |
389 |
|
o Non-Hydrostatic pickup file Pb fixed. |
390 |
|
o bug with Rigid-lid version (since c35) fixed. |
391 |
|
|
392 |
|
checkpoint36 |
393 |
|
o implement Crank-Nickelson barotropic time-stepping |
394 |
|
o recover Time-Average diagnostics |
395 |
|
o fix Pb in checkpoint (PR#19) |
396 |
|
o AIM become a standard package ; clean unused CPP_kees |
397 |
|
|
398 |
|
checkpoint35 |
399 |
|
o subtantial rewrite of dynamics.f |
400 |
|
o allows staggered time-stepping |
401 |
|
o packaged obcs, aim (Atmospheric Physics), zonal_filt, shap_filt |
402 |
|
o added bottom drag (linear + quadratic) |
403 |
|
|
404 |
|
checkpoint34 |
405 |
|
o Modified exchange routines for correct reverse simulation. |
406 |
|
o Added verification output to comply with bug fix in impldiff.F |
407 |
|
as of c33 (exp2, natl_box). |
408 |
|
o Corrected store directives in dynamics.F |
409 |
|
o Corrected array boundaries in impldiff.F |
410 |
|
o Corrected array dimensions in gmredi. |
411 |
|
Added temporary scalars to avoid storage by TAMC (gmredi, kpp). |
412 |
|
o Added routine GLOBAL_SUM_INT to global_sum.F |
413 |
|
o Added mdsio_gl.F for ECCO purposes. |
414 |
|
|
415 |
|
checkpoint33 |
416 |
|
o Fixed bug in lower BC in impldiff.F |
417 |
|
o Fixed confusion about units of forcing arrays in FFIELDS.h |
418 |
|
namely Fu,Fv,Qnet,Qsw,EmPmR: |
419 |
|
- Removed verification/natl_box/code/external_fields_scale.F |
420 |
|
(did not differ from that in model/src) |
421 |
|
- Changed units of fu,fv,Qnet,Qsw,EmPmR back to proper units |
422 |
|
(see FFIELDS.h for description) |
423 |
|
- Scale fu,fv,Qnet,Qsw,EmPmR when used in external_forcing_surf.F, |
424 |
|
kpp_calc.F and kpp_transport_t.F |
425 |
|
- Removed model/src/external_fields_scale.F and calls to it |
426 |
|
- verification/natl_box uses flux data with "atmospheric" sign so |
427 |
|
a special version of external_fields_load.F is used to |
428 |
|
change the data as it's read in. This way, the arrays |
429 |
|
have the right units and signs at all times tha a user could |
430 |
|
possibly use them. |
431 |
|
o Corrected genmake to delete a temporary file during configuration |
432 |
|
|
433 |
|
checkpoint32 |
434 |
|
o Scaling of forcing fields moved from external_forcing.F to |
435 |
|
external_fields_scale.F, called directly after loading fields. |
436 |
|
o Surface relaxation terms added to surface tendency fields in |
437 |
|
extra subroutine. |
438 |
|
o impldiff.F changed to avoid storing by TAMC of huge intermediate |
439 |
|
fields. |
440 |
|
o Changes in GMRedi to comply with packaging structure. |
441 |
|
o Modifications and bug fixes in KPP package. |
442 |
|
o Verification experiment for KPP added (verification/nat_box/) |
443 |
|
|
444 |
|
checkpoint31 |
445 |
|
o Packages interface brought in line with defined standard |
446 |
|
-> new routines packages_...F |
447 |
|
-> changes in initialise_fixed.F, initialise_varia.F |
448 |
|
-> enables initialisation interface for ECCO package |
449 |
|
o Rescaling of external forcing fields immediately after read |
450 |
|
(removed from external_forcing.F) |
451 |
|
-> new: surface_fields_scale.F |
452 |
|
-> affects ini_foorcing.F, external_fields_load.F |
453 |
|
o Computation of surface tendencies before tendency updates. |
454 |
|
-> new: external_forcing_surf.F |
455 |
|
o Change of variable names useGMRedi, useKPP, useECCO |
456 |
|
o Further changes in dynamics.F, impldiff.F for TAMC |
457 |
|
o Tested for experiments 0, 2, 4. |
458 |
|
- not yet tested for GMRedi. |
459 |
|
- not yet tested for KPP. |
460 |
|
|
461 |
|
checkpoint30 |
462 |
|
o Updates for OBCs |
463 |
|
o New experiment for test OBCs (internal_wave) |
464 |
|
o New output.txt in verification/exp4/results/ |
465 |
|
o Updates for KPP: calculation of viscosity corrected, new variables |
466 |
|
kpp_dumpFreq and kpp_taveFreq |
467 |
|
o CPP'd calls to system routines for TARGET_CRAY_VECTOR (P.H.) |
468 |
|
and entry for T90 in genmake |
469 |
|
|
470 |
|
checkpoint29 |
471 |
|
o packaged KPP and GM/Redi |
472 |
|
|
473 |
|
checkpoint28 |
474 |
|
o Corrected initialization of hFacMin for difference vertical coordinates |
475 |
|
o Modified calculation of partial cell thickness to more robust algorithm |
476 |
|
o config_summary.F: replaced write_1D_... by write_0D_... |
477 |
|
read_write.F: added subroutines write_0D_... |
478 |
|
o routines calc_common_factors, calc_gs, calc_gt, calc_phi_hyd, find_rho: |
479 |
|
included some initialisations required for TAMC. |
480 |
|
o routine calc_isoslopes: modified sqrt computation to make it |
481 |
|
differentiable for zero argument. |
482 |
|
o routines dynamics, impldiff: |
483 |
|
included store directives and tape key computations required for TAMC. |
484 |
|
o ALL TAMC related changes are between ifdef ALLOW_AUTODIFF_TAMC |
485 |
|
|
486 |
|
checkpoint27 |
487 |
|
o fixed calc_isoslopes() to so that cancellation of terms in tensor |
488 |
|
formulation works properly with variable resolution. |
489 |
|
o restructured the calling tree between the_model_main() and dynamics() |
490 |
|
o split initialise() into initialise_fixed() and initialise_varia() |
491 |
|
o introduced initialization file pSurfInitFile for free surface (cg2d_x) |
492 |
|
in new routine ini_psurf() |
493 |
|
|
494 |
|
checkpoint26 |
495 |
|
o moved some #ifdef in optional routines to encompass everything |
496 |
|
except the argument declarations. This was using up some memory. |
497 |
|
o fixed macros like _ADT to work with both versions of Linux cpp |
498 |
|
o cleaned up some unused and uninitialized variables |
499 |
|
(helps when debugging with strict compile options) |
500 |
|
o split up CPP_EEOPTIONS.h into options and macros |
501 |
|
(macros are now in eesupp/inc/CPP_EEMACROS.h) |
502 |
|
o patch for 2 processor JAM mode |
503 |
|
|
504 |
|
checkpoint25 |
505 |
|
o updates for OBCs and NH |
506 |
|
o dramatically reduced memory usage |
507 |
|
o added "JAM" routines for use on Artic network |
508 |
|
o parameterization of convection by implicit vertical diffusion |
509 |
|
o added a diagnostic of convective index |
510 |
|
o moved call to calc_divg_hat() from dynamics() to solve_for_pressure() |
511 |
|
(this is was partly for OBCs and NH but also in anticipation of |
512 |
|
an explicit free-surface option) |
513 |
|
o initial state input files for U and V |
514 |
|
|
515 |
|
checkpoint24 |
516 |
|
o Removed some spurious "REAL"s |
517 |
|
o Updated KPP to latest version from JPL |
518 |
|
o Added a COS(latitude) in horizontal viscosity |
519 |
|
|
520 |
|
checkpoint23 |
521 |
|
o Added del^4 diffusion for salt and temperature |
522 |
|
o Multiple minor fixes (implicit none's, arguments to global_sum, etc.) |
523 |
|
|
524 |
|
checkpoint22 |
525 |
|
o Made many small changes: implicit none, format(i5), etc... |
526 |
|
o Introduced KPP mixing scheme (compiles and runs but *not* tested) |
527 |
|
|
528 |
|
checkpoint21 |
529 |
|
o Introduced new I/O package that supports direct-access records (mdsio.F) |
530 |
|
o Split-up mid-level I/O routines |
531 |
|
o Updated input data for experiments 0,1,2,4 and 5 |
532 |
|
o Finally removed the genmake.* scripts (everyone seems happy with genmake) |
533 |
|
|
534 |
|
checkpoint20 |
535 |
|
o Tidied up open-boundaries some more |
536 |
|
o Added non-hydrostatic code |
537 |
|
o Fixed some input data |
538 |
|
o Modified MPI start-up to work in "coupled" context. |
539 |
|
|
540 |
|
checkpoint19 |
541 |
|
o Tidied up directory structure for verification experiments. |
542 |
|
o Added new code for open boundaries. |
543 |
|
o Corrected advective terms near boundaries. |
544 |
|
o Added CPP flag for "old UV" geometry and corrected spherical geometry. |
545 |
|
o Implemented free-slip/no-slip BCs and del^4 dissipation. |
546 |
|
o Split default parameters out of ini_parms.F into set_defaults.F. |
547 |
|
|
548 |
|
checkpoint18 |
549 |
|
o Further changes for optional code generation with |
550 |
|
CPP - ugh! |
551 |
|
|
552 |
|
checkpoint17 |
553 |
|
o Beginning to incorporating modularity |
554 |
|
for supporting atmos. and ocean configs. |
555 |
|
o Small changes for TAMC complinace |
556 |
|
o New ultra simple test case. Barotropic box |
557 |
|
configuration "exp0". |
558 |
|
|
559 |
|
checkpoint16 |
560 |
|
o Minor changes to exchange routines to support |
561 |
|
accumulation in reverse mode. Adjoint of an |
562 |
|
assignment is an addition. |
563 |
|
o Changes to support compilation by TAMC (LR(1) |
564 |
|
syntax) and by g77 under Linux ( 72 character |
565 |
|
limit ). |
566 |
|
|
567 |
|
checkpoint15 |
568 |
|
Rewrote exchange routines to allow |
569 |
|
o General tile <-> tile connectivity. |
570 |
|
o TAMC reverse mode flag ( is this really needed? ) |
571 |
|
o Variable width overlap and exchange region widths. |
572 |
|
o Hooks for sharing data via DMA, Arctic, Memory Channel, |
573 |
|
shmput/shmget, VIA, SYSV shmem and every other cool |
574 |
|
communication method. |
575 |
|
|
576 |
|
checkpoint14 |
577 |
|
Consistent isomorphism chages made. These allow dynamical core |
578 |
|
to support both pressure and meters as a vertical coordinate. |
579 |
|
|
580 |
|
checkpoint13 |
581 |
|
Release which corrects global_max problem and a couple |
582 |
|
of KGM time-averaging diagnostic errors. Also contains |
583 |
|
a double-gyre experiment setup as well as the baseline |
584 |
|
4 degree global template. The double gyre template is |
585 |
|
organised to allow it to be applied automatically to |
586 |
|
the baseline configuration. |
587 |
|
|
588 |
|
branch-point-rdot |
589 |
|
A few tidy-ups have been made bt the real point of this |
590 |
|
check-point (excuse the pun) is to allow a branch to be |
591 |
|
made for the purposes of adding the rDot alterations. |
592 |
|
I don't want to commit the rDot to the main-trunk until |
593 |
|
we've actually tested it... |
594 |
|
|
595 |
checkpoint12 |
checkpoint12 |
596 |
Tidy up and added generic genmake script (formerly |
Tidy up and added generic genmake script (formerly |
597 |
configure script). |
configure script). |