2 |
|
|
3 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
4 |
============================== |
============================== |
5 |
|
checkpoint45d_post |
6 |
|
o import 2 fixes from release1_p5 (CG2D.h & packages_readparms.F) |
7 |
|
o use recip_dx*,recip_dy* instead of /dx*,/dy* in orlanski_E,W,N,S |
8 |
|
(affects plume_on_slope "cg2d" output) |
9 |
|
o add new parameter deltaTfreesurf for free-surface time stepping |
10 |
|
o exchange gT,gS when using staggerTimeStep & GM & Oly<4 |
11 |
|
o NONLIN_FRSURF: rescale gNm1 to get a better conservation with A-B |
12 |
|
o change phiMin in exp1 (to agree with documentation) |
13 |
|
o add new flags: |
14 |
|
- T,S forcing inside/outside Adams-Bashforth |
15 |
|
- turn off temp,salt advection and/or forcing |
16 |
|
- for each tracer, define internal flag for multiDimAdvection & A-B |
17 |
|
o monitor: change definition of KE to check conservation of Energy |
18 |
|
|
19 |
|
checkpoint45c_post |
20 |
|
o Removed f90 comments ("!") |
21 |
|
o Modified initialisations to enable exf package for MITgcm |
22 |
|
(without invoking ECCO_PACKAGE). |
23 |
|
o modifications to gradient check package (Martin Losch) |
24 |
|
- enable centered differences |
25 |
|
- modified format of standard output |
26 |
|
|
27 |
|
checkpoint45b_post |
28 |
|
o cleaned exchange calls in cg2d |
29 |
|
o Included CPP option ALLOW_AUTODIFF_MONITOR in forward_step |
30 |
|
o included CPP option SINGLE_LAYER_MODE in thermodynamics |
31 |
|
to configure barotropic setup (Martin Losch) |
32 |
|
o moved some initialisations within bi,bj loop in dynamics |
33 |
|
(TAF flow dependency) |
34 |
|
o in initialise_varia, moved call packages_init_variables |
35 |
|
before convective_adjustment_init |
36 |
|
(corresponding adjustment of TAF tape initialisation). |
37 |
|
o in calc_diffusivity/calc_viscosity extended domain to full overlap. |
38 |
|
|
39 |
|
checkpoint45a_post |
40 |
|
o compute Rho over the whole domain (extend iMin,jMin indices) |
41 |
|
o specific exchange for X-slice domain (case Ny=1) |
42 |
|
o argument futureIter of S/R obcs_calc missing in exp4 & internal_wave /code |
43 |
|
o solve_for_pressure : set cg2d_x to zero where OBCS are applied |
44 |
|
|
45 |
|
checkpoint44h_post = checkpoint45 |
46 |
|
o Fix to restore differentiability: |
47 |
|
place at which packages_init_variables is called in |
48 |
|
initialise_varia is crucial for reverse control flow. |
49 |
|
o modifications in GMRedi to enable stable adjoint |
50 |
|
(corresponding modif's in thermodynamics, dynamics) |
51 |
|
o added missing hook for sponge layer code in external_forcing |
52 |
|
o modified test cost functions |
53 |
|
o storing in gad_advection (plus removal of write(0,*)! |
54 |
|
o missing headers TR1.h added in convective_adjustment routines |
55 |
|
o errorMessageUnit set non-equal zero in eeboot |
56 |
|
(conflict on some platforms, e.g. IBM SP3) |
57 |
|
o modified carbon verif. |
58 |
|
|
59 |
|
checkpoint44h_pre |
60 |
|
o vertical grid option: allow to put the Interface at the middle between |
61 |
|
2 cell-centers ; replace delZ in pkg/kpp by drF. |
62 |
|
o GM Advective form: Tracers are advected using the residual transport (= |
63 |
|
Euler+GM-bolus); set param GM_AdvSeparate=T to return to previous form. |
64 |
|
|
65 |
|
checkpoint44g_post |
66 |
|
o fix surface correction term in multiDimAdvection (affects plume_on_slope |
67 |
|
results) ; ifdef missing in thermodynamics.F |
68 |
|
|
69 |
|
checkpoint44f_pre,checkpoint44f_post |
70 |
|
o added PTRACERS package |
71 |
|
This allows an arbitrary number of passive tracers to be integrated |
72 |
|
forward simultaneously with the dynamicaly model. |
73 |
|
+ Implemented so far: |
74 |
|
- basic forward algorithm (time-stepping, advection, diffusion, convection) |
75 |
|
- I/O and checkpointing |
76 |
|
- GM/Redi *but* using the GM/Redi coefficient of Salt |
77 |
|
+ Not implemented so far: |
78 |
|
- KPP |
79 |
|
- OBCS |
80 |
|
+ No specific example supplied (yet) but global_ocean.90x40x15 has the |
81 |
|
necessary data.ptracer file. Simply use -enable=ptracers and uncomment |
82 |
|
line in data.pkg. |
83 |
|
+ This package is disabled by default since it increases storage. |
84 |
|
|
85 |
|
|
86 |
|
o testscript: added -cleanup option |
87 |
|
o Shap_filt: a) new shap_filt S/R to use no-slip BC with S2 filter |
88 |
|
b) enable to filter 2D fields. |
89 |
|
|
90 |
|
checkpoint44e_pre,post |
91 |
|
o re-wrote convective adjustment to use pre-calculated weights |
92 |
|
Affects: model/src/convective_adjustment.F, convective_adjustment_ini.F |
93 |
|
Adds: model/src/convective_weights.F, convectively_mixtracer.F |
94 |
|
Original calls to convect() still in place. |
95 |
|
|
96 |
|
chkpt44d_pre,post |
97 |
|
o added missing EXCLUDE_MONITOR flags |
98 |
|
o changed "e" to "_d" in gmredi_slope_limit, gmredi_slope_psi |
99 |
|
(incompatible typ in MIN/MAX expressions caused problems |
100 |
|
on IBM SP3) |
101 |
|
o in genmake added variable MAKEDEPEND |
102 |
|
plus resetting for case SunOS |
103 |
|
o added timer_stats.c routine for IBM SP3 |
104 |
|
o removed variables in dynamics |
105 |
|
!!! I forgot to create tag chkpt44d_pre !!! |
106 |
|
!!! But chkpt44d_pre = chkpt44c_post !!! |
107 |
|
|
108 |
|
o real fresh water flux implemented with non-linear free-surface. |
109 |
|
|
110 |
|
chkpt44c_pre,post |
111 |
|
o few fix (mask in shap_s2, EmPmR in external_field_load, |
112 |
|
USE_NATURAL_BCS in solve_for_P); |
113 |
|
o add arguments myIter & myTime to S/R obcs_calc & solve_for_P |
114 |
|
|
115 |
|
checkpoint44b_pre/_post |
116 |
|
o merge of relevant stuff from the ecco-branch: |
117 |
|
- genmake: removed $S64 overwrite for case SunOS |
118 |
|
- pkg/exf: update and corrections for field swapping and obcs |
119 |
|
- pkg/ecco: parameter lists for the_model_main, the_main_loop |
120 |
|
harmonized between ECCO and MITgcm |
121 |
|
- pkg/autodiff: added flow directives for obcs, mdsio_gl_slice |
122 |
|
updated checkpointing_lev... lists for obcs |
123 |
|
- model/src: minor changes in forward_step, plot_field |
124 |
|
added directive for divided adjoint in the_main_loop |
125 |
|
- pkg/mdsio: added mdsio_gl_slice |
126 |
|
|
127 |
|
o check parameters & config (chkpt44a_pre,post) |
128 |
|
o OBC and NonLin_FrSurf. |
129 |
|
|
130 |
|
checkpoint44 |
131 |
|
o fix bug in mom_vi_del2uv |
132 |
|
o select when filters are applied ; add options to zonal_filter (data.zonfilt) |
133 |
|
o gmredi: fix Pb in the adiabatic form ; add options (.e.g. Bolus advection) |
134 |
|
o update AIM experiments (NCEP input files) |
135 |
|
o improve and extend diagnostics (Monitor, TimeAve with NonLin-FrSurf) |
136 |
|
o added some stuff for AD |
137 |
|
|
138 |
|
checkpoint43 |
139 |
|
o added two new advection tests |
140 |
|
o added global 4x4 degree ocean test |
141 |
|
o added test of H&S94 with free surface on cube |
142 |
|
o some final mods for gradient check, adjoint,... |
143 |
|
|
144 |
|
checkpoint42 |
145 |
|
Fixed AD-related problems: |
146 |
|
o Store directives up-to-date with re-arranged Adams-Bashforth |
147 |
|
(mainly thermodynamics.F) |
148 |
|
o New store directives for multi-dim. advection schemes |
149 |
|
* new CPP flag ALLOW_MULTI_DIM_ADVECTION |
150 |
|
* new common block and key passkey |
151 |
|
(mainly gad_advection.F) |
152 |
|
o Modified store directives for split of dynamics/thermodynamics |
153 |
|
for the case ALLOW_KPP |
154 |
|
o Cleaned argument list for timestep_tracer.F |
155 |
|
|
156 |
|
checkpoint41 |
157 |
|
o re-formatted/added comments in prologues of key subroutines |
158 |
|
to work with protex |
159 |
|
|
160 |
|
checkpoint40 |
161 |
|
o moved calc_gs, calc_gt and calc_tr1 back to model/src |
162 |
|
o added FLT (floats) package from Arne along with demo. experiment |
163 |
|
o re-arranged Adams-Bashforth scheme for tracers to allow easier |
164 |
|
implementation of forward-in-time forcing terms |
165 |
|
o more mods for non-linear free-surface including new variable |
166 |
|
o modified multi-dim method to work on cube |
167 |
|
o bug fix in KPP (from DM) |
168 |
|
|
169 |
|
checkpoint40pre9 |
170 |
|
o added DST advection schemes |
171 |
|
o fix MPI bug on SGI's |
172 |
|
o modifications for non-linear free-surface |
173 |
|
o fixed conditionals for fixed flow/passive tracer only mode |
174 |
|
o added tension/strain form of stress tensor |
175 |
|
o changes to commenting of arguments for TAF/TAMC |
176 |
|
o added/updated many matlab scripts |
177 |
|
o added multi-dimensional advection method for tracers |
178 |
|
|
179 |
|
checkpoint40pre8 |
180 |
|
o new package rw/ |
181 |
|
Added method for dumping intermediate local arrays: |
182 |
|
mdsio_writetile - same as mdsio_writefield except works from inside |
183 |
|
bi,bj loop |
184 |
|
mdsio_writelocal - same as mdsio_writetile except works for local |
185 |
|
arrays |
186 |
|
write_local_r? - higher-level wrapper for mdsio_writelocal |
187 |
|
Controlled by diagFreq. Defaults to zero (ie. no dumps) |
188 |
|
Example given at end of mom_vecinv.F that dumps some local arrays. |
189 |
|
o replaced calc_mom_rhs.F by mom_fluxform.F/mom_vecinc.F to |
190 |
|
control flux-form versus vector invariant code. |
191 |
|
Switch is runtime flag vectorInvariantMomentum (defaults to false) |
192 |
|
or genmake -DISABLE ... |
193 |
|
o TAMC-related update to cope with mom_fluxform.F/mom_vecinc.F |
194 |
|
(additional storing is now needed for case vecinv; |
195 |
|
currently nothing implemented to switch this storing off |
196 |
|
in case fluxform is used since no CPP flag available). |
197 |
|
|
198 |
|
checkpoint40pre7 |
199 |
|
o bug fix in dynamics.F |
200 |
|
calc_viscosity called before k=1,Nr loop. |
201 |
|
o added some #ifdef's ALLOW_PASSIVE_TRACER |
202 |
|
o cleaned initialisations in ctrl_map_ini |
203 |
|
related to TAF |
204 |
|
|
205 |
|
checkpoint40pre6 |
206 |
|
o split dynamics into thermodynamics & dynamics |
207 |
|
o corresponding split of calc_diffusivity into |
208 |
|
calc_diffusivity, calc_viscosity |
209 |
|
(kpp_calc_diff, kpp_calc_visc). |
210 |
|
o Added diffkr, kapgm to control variables |
211 |
|
o bug fix in gmredi_slope_limit for ldd97 |
212 |
|
|
213 |
|
checkpoint40pre5 |
214 |
|
o proper initialization of Adams-Bashforth time stepping |
215 |
|
+ uses forward step for very first step of model (iter=0) |
216 |
|
o re-generated *all* output since almost all output is modified by this |
217 |
|
change (sorry but we've been putting this one off for about 2 years) |
218 |
|
+ natl_box, aim.5l_Equatorial_Channel and aim.5l_zon-ave were |
219 |
|
*not* regenerated since they are already out of date and I don't |
220 |
|
want to create the impression that they are suddenly working. |
221 |
|
o removed KAP stuff for DEC's |
222 |
|
o modified/added many .cvsignore files in */code and */input |
223 |
|
o added new expt "solid-body.cs-32x32x1" |
224 |
|
+ this uses an alternative form of ini_curvilinear_grid.F which |
225 |
|
is not necessarily any better than the current one... |
226 |
|
|
227 |
|
checkpoint40pre4 |
228 |
|
o added re-initialisations of utrans,vtrans,rtrans in |
229 |
|
calc_common_factors which are crucial for TAF. |
230 |
|
o extended iMin,jMin range for calc_common_factors, |
231 |
|
calc_diffusivity |
232 |
|
o added option ALLOW_PASSIVE_TRACER |
233 |
|
o a few minor updates and fixes |
234 |
|
|
235 |
|
checkpoint40pre3 |
236 |
|
o created calc_gtr1.F for gad package |
237 |
|
o split the_main_loop and re-introduced forward_step |
238 |
|
o some recomputations of nIter0 for TAMC |
239 |
|
|
240 |
|
checkpoint40pre2 |
241 |
|
o Updated to latest branch of AJA, JMC |
242 |
|
o Added gradient check package |
243 |
|
o Added routines for passive tracer handling |
244 |
|
o Added carbon sequestration sensitivity experiment |
245 |
|
|
246 |
|
checkpoint39 |
247 |
|
o Modifications and fixes to support TAMC |
248 |
|
(mainly missing or wrong directives) |
249 |
|
o Corrected common blocks when fooling TAMC to produce adjoint I/O |
250 |
|
in addummy_in_stepping |
251 |
|
o Updated verification/vero/ (bug fixes in code/, correct pickups in input/) |
252 |
|
o Enabled "sliced" (xz-,yz-) exchanges, needed for obcs. |
253 |
|
o Included following packages: |
254 |
|
cal: calendar |
255 |
|
exf: external forcing |
256 |
|
ecco: essential routines to control use of cal, exf |
257 |
|
(NB: this is not the full ECCO environment) |
258 |
|
o disabled some packages by default in genmake |
259 |
|
|
260 |
|
checkpoint38 |
261 |
|
o see notes_c37_adj.txt |
262 |
|
o corrected wrong layer index in external_forcing_t |
263 |
|
o replaced the_main_loop/forward_step by the_main_loop |
264 |
|
o removed forceoutput from write_state |
265 |
|
(replaced criteria for initial/final state) |
266 |
|
o added runtime flag for pre-C35 pickups: usePickupBeforeC35 |
267 |
|
|
268 |
|
checkpoint37 |
269 |
|
o mdsio.F routine split and packaged in pkg/mdsio |
270 |
|
(mdsio.F and mdsio_gl.F still exist in eesupp/src but are not |
271 |
|
used unless genmake -disable=mdsio) |
272 |
|
o allow a more precise surface-geopotential gradient for the atmosphere |
273 |
|
taking into account the change in space of Alpha (=1/rho) at the surface. |
274 |
|
o variable "etaN" (replace cg2d_x) is the surface R-anomaly [r unit] |
275 |
|
(ocean: sea surface height [m] / atmos: surface pressure anomaly [Pa]) |
276 |
|
o Change units: Phi_Hydrostatic (=phiHyd), Phi_Surface(=cg2d_x) and |
277 |
|
Phi_NonHydrostatic (=cg3d_x), all have now the dimension of a potential |
278 |
|
[m2/s2] (pressure/rhoConst in the ocean ; Geo-Potential in the atmosphere). |
279 |
|
o Time-Average diagnostics defined as a pseudo package "timeave". |
280 |
|
o pickup files : old pickup (before c35) did not work with new code (c35 |
281 |
|
and after) unless activating 2 commented lines in initialise-varia. |
282 |
|
o Non-Hydrostatic pickup file Pb fixed. |
283 |
|
o bug with Rigid-lid version (since c35) fixed. |
284 |
|
|
285 |
|
checkpoint36 |
286 |
|
o implement Crank-Nickelson barotropic time-stepping |
287 |
|
o recover Time-Average diagnostics |
288 |
|
o fix Pb in checkpoint (PR#19) |
289 |
|
o AIM become a standard package ; clean unused CPP_kees |
290 |
|
|
291 |
|
checkpoint35 |
292 |
|
o subtantial rewrite of dynamics.f |
293 |
|
o allows staggered time-stepping |
294 |
|
o packaged obcs, aim (Atmospheric Physics), zonal_filt, shap_filt |
295 |
|
o added bottom drag (linear + quadratic) |
296 |
|
|
297 |
|
checkpoint34 |
298 |
|
o Modified exchange routines for correct reverse simulation. |
299 |
|
o Added verification output to comply with bug fix in impldiff.F |
300 |
|
as of c33 (exp2, natl_box). |
301 |
|
o Corrected store directives in dynamics.F |
302 |
|
o Corrected array boundaries in impldiff.F |
303 |
|
o Corrected array dimensions in gmredi. |
304 |
|
Added temporary scalars to avoid storage by TAMC (gmredi, kpp). |
305 |
|
o Added routine GLOBAL_SUM_INT to global_sum.F |
306 |
|
o Added mdsio_gl.F for ECCO purposes. |
307 |
|
|
308 |
|
checkpoint33 |
309 |
|
o Fixed bug in lower BC in impldiff.F |
310 |
|
o Fixed confusion about units of forcing arrays in FFIELDS.h |
311 |
|
namely Fu,Fv,Qnet,Qsw,EmPmR: |
312 |
|
- Removed verification/natl_box/code/external_fields_scale.F |
313 |
|
(did not differ from that in model/src) |
314 |
|
- Changed units of fu,fv,Qnet,Qsw,EmPmR back to proper units |
315 |
|
(see FFIELDS.h for description) |
316 |
|
- Scale fu,fv,Qnet,Qsw,EmPmR when used in external_forcing_surf.F, |
317 |
|
kpp_calc.F and kpp_transport_t.F |
318 |
|
- Removed model/src/external_fields_scale.F and calls to it |
319 |
|
- verification/natl_box uses flux data with "atmospheric" sign so |
320 |
|
a special version of external_fields_load.F is used to |
321 |
|
change the data as it's read in. This way, the arrays |
322 |
|
have the right units and signs at all times tha a user could |
323 |
|
possibly use them. |
324 |
|
o Corrected genmake to delete a temporary file during configuration |
325 |
|
|
326 |
|
checkpoint32 |
327 |
|
o Scaling of forcing fields moved from external_forcing.F to |
328 |
|
external_fields_scale.F, called directly after loading fields. |
329 |
|
o Surface relaxation terms added to surface tendency fields in |
330 |
|
extra subroutine. |
331 |
|
o impldiff.F changed to avoid storing by TAMC of huge intermediate |
332 |
|
fields. |
333 |
|
o Changes in GMRedi to comply with packaging structure. |
334 |
|
o Modifications and bug fixes in KPP package. |
335 |
|
o Verification experiment for KPP added (verification/nat_box/) |
336 |
|
|
337 |
|
checkpoint31 |
338 |
|
o Packages interface brought in line with defined standard |
339 |
|
-> new routines packages_...F |
340 |
|
-> changes in initialise_fixed.F, initialise_varia.F |
341 |
|
-> enables initialisation interface for ECCO package |
342 |
|
o Rescaling of external forcing fields immediately after read |
343 |
|
(removed from external_forcing.F) |
344 |
|
-> new: surface_fields_scale.F |
345 |
|
-> affects ini_foorcing.F, external_fields_load.F |
346 |
|
o Computation of surface tendencies before tendency updates. |
347 |
|
-> new: external_forcing_surf.F |
348 |
|
o Change of variable names useGMRedi, useKPP, useECCO |
349 |
|
o Further changes in dynamics.F, impldiff.F for TAMC |
350 |
|
o Tested for experiments 0, 2, 4. |
351 |
|
- not yet tested for GMRedi. |
352 |
|
- not yet tested for KPP. |
353 |
|
|
354 |
|
checkpoint30 |
355 |
|
o Updates for OBCs |
356 |
|
o New experiment for test OBCs (internal_wave) |
357 |
|
o New output.txt in verification/exp4/results/ |
358 |
|
o Updates for KPP: calculation of viscosity corrected, new variables |
359 |
|
kpp_dumpFreq and kpp_taveFreq |
360 |
|
o CPP'd calls to system routines for TARGET_CRAY_VECTOR (P.H.) |
361 |
|
and entry for T90 in genmake |
362 |
|
|
363 |
|
checkpoint29 |
364 |
|
o packaged KPP and GM/Redi |
365 |
|
|
366 |
|
checkpoint28 |
367 |
|
o Corrected initialization of hFacMin for difference vertical coordinates |
368 |
|
o Modified calculation of partial cell thickness to more robust algorithm |
369 |
|
o config_summary.F: replaced write_1D_... by write_0D_... |
370 |
|
read_write.F: added subroutines write_0D_... |
371 |
|
o routines calc_common_factors, calc_gs, calc_gt, calc_phi_hyd, find_rho: |
372 |
|
included some initialisations required for TAMC. |
373 |
|
o routine calc_isoslopes: modified sqrt computation to make it |
374 |
|
differentiable for zero argument. |
375 |
|
o routines dynamics, impldiff: |
376 |
|
included store directives and tape key computations required for TAMC. |
377 |
|
o ALL TAMC related changes are between ifdef ALLOW_AUTODIFF_TAMC |
378 |
|
|
379 |
|
checkpoint27 |
380 |
|
o fixed calc_isoslopes() to so that cancellation of terms in tensor |
381 |
|
formulation works properly with variable resolution. |
382 |
|
o restructured the calling tree between the_model_main() and dynamics() |
383 |
|
o split initialise() into initialise_fixed() and initialise_varia() |
384 |
|
o introduced initialization file pSurfInitFile for free surface (cg2d_x) |
385 |
|
in new routine ini_psurf() |
386 |
|
|
387 |
|
checkpoint26 |
388 |
|
o moved some #ifdef in optional routines to encompass everything |
389 |
|
except the argument declarations. This was using up some memory. |
390 |
|
o fixed macros like _ADT to work with both versions of Linux cpp |
391 |
|
o cleaned up some unused and uninitialized variables |
392 |
|
(helps when debugging with strict compile options) |
393 |
|
o split up CPP_EEOPTIONS.h into options and macros |
394 |
|
(macros are now in eesupp/inc/CPP_EEMACROS.h) |
395 |
|
o patch for 2 processor JAM mode |
396 |
|
|
397 |
|
checkpoint25 |
398 |
|
o updates for OBCs and NH |
399 |
|
o dramatically reduced memory usage |
400 |
|
o added "JAM" routines for use on Artic network |
401 |
|
o parameterization of convection by implicit vertical diffusion |
402 |
|
o added a diagnostic of convective index |
403 |
|
o moved call to calc_divg_hat() from dynamics() to solve_for_pressure() |
404 |
|
(this is was partly for OBCs and NH but also in anticipation of |
405 |
|
an explicit free-surface option) |
406 |
|
o initial state input files for U and V |
407 |
|
|
408 |
|
checkpoint24 |
409 |
|
o Removed some spurious "REAL"s |
410 |
|
o Updated KPP to latest version from JPL |
411 |
|
o Added a COS(latitude) in horizontal viscosity |
412 |
|
|
413 |
|
checkpoint23 |
414 |
|
o Added del^4 diffusion for salt and temperature |
415 |
|
o Multiple minor fixes (implicit none's, arguments to global_sum, etc.) |
416 |
|
|
417 |
|
checkpoint22 |
418 |
|
o Made many small changes: implicit none, format(i5), etc... |
419 |
|
o Introduced KPP mixing scheme (compiles and runs but *not* tested) |
420 |
|
|
421 |
checkpoint21 |
checkpoint21 |
422 |
o Introduced new I/O package that supports direct-access records (mdsio.F) |
o Introduced new I/O package that supports direct-access records (mdsio.F) |