1 |
$Header: /u/gcmpack/MITgcm/doc/tag-index,v 1.118 2003/01/13 14:31:49 adcroft Exp $ |
2 |
|
3 |
Notes on tags used in MITgcmUV |
4 |
============================== |
5 |
|
6 |
checkpoint47i_post |
7 |
o GM_EXCLUDE_() replace GM_TAPER_(): by default, all the code is compiled. |
8 |
o add GM_Small_Number & GM_slopeSqCutoff as run-time parameters (data.gmredi) |
9 |
default values = 1.D-12 & 1.D+48 (=> give same results as checkpoint47f_post) |
10 |
o cg2dTargetResWunit: define tolerance using deltaTmom (and not deltaTfreesurf) |
11 |
o gmredi: fix few bugs. |
12 |
|
13 |
checkpoint47h_post (duplicate tag for PH's c48 tag, in order to move c48) |
14 |
This will be a good checkpoint to plug into the ecco-branch. |
15 |
o few modif.'s to get latest version adjointed |
16 |
(mainly kick out code in ini_linear_phisurf) |
17 |
o modif's to run adjoint with exactConserv |
18 |
o case GM_BOLUS_ADVEC should be cleaned |
19 |
S/R gmredi_slope_psi should be cleaned |
20 |
o verification/carbon now has exactConserv=.TRUE. |
21 |
|
22 |
checkpoint47g_post |
23 |
o time-average AIM output: becomes independent from statvars time-average |
24 |
(can be used with #undef ALLOW_TIMEAVE) ; add new variables (e.g., RH) |
25 |
o After TAF bug fix: |
26 |
Tested adjoint of gmredi package. |
27 |
Tests were done with checkpoint47 rather than with latest |
28 |
checkpoint47f_post since the latter is broken for the adjoint |
29 |
verification/carbon/ contains both ADM and TLM gradient checks. |
30 |
Here they are, before they get changed/lost/stolen. |
31 |
Mostly modif.'s to fix numerical sensitivities. |
32 |
Gradient checks OK for |
33 |
- GM_taper_scheme: |
34 |
* clipping |
35 |
* ac02 |
36 |
* linear |
37 |
* glw91 |
38 |
* dm95 |
39 |
* ldd97 |
40 |
- GMREDI_OPTIONS: |
41 |
* GM_VISBECK_VARIABLE_K |
42 |
* GM_NON_UNITY_DIAGONAL |
43 |
* GM_EXTRA_DIAGONAL |
44 |
* GM_BOLUS_ADVEC |
45 |
in conjunction with data.gmredi parameters to be checked in |
46 |
in a few minutes under verification/carbon/code/ |
47 |
|
48 |
checkpoint47f_post |
49 |
Merging from release1_p10: |
50 |
o modifications for using pkg/exf with pkg/seaice |
51 |
- pkg/seaice CPP options SEAICE_EXTERNAL_FORCING |
52 |
and SEAICE_EXTERNAL_FLUXES |
53 |
- pkg/exf CPP options EXF_READ_EVAP and |
54 |
EXF_NO_BULK_COMPUTATIONS |
55 |
- usage examples are Experiments 8 and 9 in |
56 |
verification/lab_sea/README |
57 |
- verification/lab_sea default experiment now uses |
58 |
pkg/gmredi, pkg/kpp, pkg/seaice, and pkg/exf |
59 |
|
60 |
checkpoint47e_post |
61 |
o created a tag prior to creating a branch |
62 |
|
63 |
checkpoint47d_post |
64 |
o fixed bug in the oceanic pressure coordinates code: vertical viscosity |
65 |
at the bottom boundary had an erroneous half slip boundary condition |
66 |
o added bottom drag and no slip boundary condition capabilities to |
67 |
oceanic pressure coordinates code |
68 |
o added new verification experiment global_ocean_pressure, this experiment is |
69 |
also described in the tutorial section of the manual |
70 |
checkpoint47d_pre |
71 |
|
72 |
checkpoint47c_post |
73 |
o Added a verification experiment for a global cubed-sphere ocean run |
74 |
- verification/global_ocean.cubed32x32x30 |
75 |
- uses pkg/therm_seaice and pkg/bulk_forcing |
76 |
|
77 |
o allow to run AIM physics with SPEEDY input files (from Franco Molteni) |
78 |
o allow a more accurate definition of Ro_Surf (selectFindRoSurf=1) |
79 |
when using P-coordinate; only implemented for atmospheric config. |
80 |
o OCEANICP & realFreshWater: include P-E direct effect on wVel ; |
81 |
NOTES: requires option NONLIN_FRSURF to be "#define". |
82 |
o update advect_xz/results/output.txt (left from checkpoint44g_post) |
83 |
|
84 |
checkpoint47b_post |
85 |
Merging from release1_p9: |
86 |
o pkg/seaice |
87 |
- removed GOTO's and added taf directives |
88 |
- double precision constants to reduce the g77 (Linux) |
89 |
to F77 (SGI) differences reported in release1_p8 |
90 |
o tools/genmake |
91 |
- added SGI options |
92 |
o verification/testscript |
93 |
- updated to that of checkpoint47a_post |
94 |
o verification/global_ocean.90x40x15/input/eedata |
95 |
- modified for SGI f77 compatibility |
96 |
o verification/lab_sea |
97 |
- added description of sea-ice model |
98 |
- added missing matlab routines |
99 |
- added test of thermodynamics parallelization |
100 |
|
101 |
checkpoint47a_post |
102 |
o new pkg: aim_v23 |
103 |
= F. Molteni atmos.physics (SPEEDY, ver23) adapted to MITgcm. |
104 |
for now, keep the same surface forcing as before. |
105 |
- Part-Cell implemented into AIM; check that Heat & Water are conserved |
106 |
- aim.5l_cs: use new aim pkg (run 10yr & get better results than before) |
107 |
- aim.5l_LatLon & aim.5l_Equatorial_Channel are still using old aim pkg |
108 |
o add diagnostic of surface correction term in monitor output |
109 |
|
110 |
o bulk_force and therm_seaice |
111 |
- Two packages: bulk_force (Bulk forcing) |
112 |
and therm_seaice (thermodynamic_seaice) - adopted from LANL CICE.v2.0.2 |
113 |
- Earlier integration from Stephaine Dutkiewicz |
114 |
and Patrick Heimbach. |
115 |
- Two ifdef statements for compile time, |
116 |
ALLOW_THERM_SEAICE and ALLOW_BULK_FORCE |
117 |
|
118 |
checkpoint47 |
119 |
differentiable version of checkpoint46n_post |
120 |
o bug fix in quasihydrostaticterms |
121 |
o * "clean" adjoint code (in terms of extensive recomputations) |
122 |
can now be obtained for all GMREDI options (i.e. for |
123 |
- GM_VISBECK_VARIABLE_K |
124 |
- GM_NON_UNITY_DIAGONAL |
125 |
- GM_EXTRA_DIAGONAL |
126 |
- GM_BOLUS_ADVEC ) |
127 |
* However, wrong gradient check problem remains unsolved. |
128 |
* New CPP options have been introduced for different |
129 |
tapering schemes |
130 |
o external_fields_load now part of differentiation list |
131 |
o pressure needs multiple storing; |
132 |
would be nice to have store_pressure at beginning or |
133 |
end of forward_step, e.g. by having phiHyd global (5-dim.) |
134 |
(NB: pressure is needed for certain cases in find_rho, |
135 |
which is also invoked through convective_adjustment). |
136 |
o recomputations in find_rho for cases |
137 |
'JMD95'/'UNESCO' or 'MDJWF' are OK. |
138 |
o #define ATMOSPHERIC_LOADING should be differentiable |
139 |
o ini_forcing shifted to begining of initialise_varia |
140 |
o Incorporating QNVS line search routines into MITgcm |
141 |
(this is separate code, and therefore not under pkg) |
142 |
lsopt/, optim/ |
143 |
o Updated verification/carbon/ |
144 |
- #define GM_VISBECK_VARIABLE_K |
145 |
- GM_taper_scheme = 'ldd97' |
146 |
|
147 |
checkpoint46n_post |
148 |
Merging from release1_p8: |
149 |
o verification/natl_box: |
150 |
updating new external_fields_load routine |
151 |
o New package: pkg/seaice |
152 |
Sea ice model by D. Menemenlis (JPL) and Jinlun Zhang (Seattle). |
153 |
The sea-ice code is based on Hibler (1979-1980). |
154 |
Two sea-ice dynamic solvers, ADI and LSR, are included. |
155 |
In addition to computing prognostic sea-ice variables and diagnosing |
156 |
the forcing/external data fields that drive the ocean model, |
157 |
SEAICE_MODEL also sets theta to the freezing point under sea-ice. |
158 |
The implied surface heat flux is then stored in variable |
159 |
surfaceTendencyTice, which is needed by KPP package (kpp_calc.F and |
160 |
kpp_transport_t.F) to diagnose surface buoyancy fluxes and for the |
161 |
non-local transport term. Because this call precedes model |
162 |
thermodynamics, temperature under sea-ice may not be "exactly" at |
163 |
the freezing point by the time theta is dumped or time-averaged. |
164 |
N.B.: fairly large differences in accuracy occur across |
165 |
different platforms/compilers; comparison between |
166 |
g77 (Linux) and F77 (SGI Origin 2000/MIPSpro) |
167 |
generated output gives: |
168 |
T S U V |
169 |
C D M c m s m s m s m s |
170 |
n p a R g m m e . m m e . m m e . m m e . |
171 |
f n k u 2 i a a d i a a d i a a d i a a d |
172 |
g d e n d n x n . n x n . n x n . n x n . |
173 |
|
174 |
Y Y Y Y 5 5 7 7 7 8 10 9 6 6 6 6 7 5 7 5 7 FAIL lab_sea |
175 |
o GAD: |
176 |
- generated new common blocks to account for call of |
177 |
same gad routines with differing traceridentities |
178 |
(needed to modify tracerIdentity indices in GAD.h) |
179 |
- generated separate common blocks for case useCubedSphereExchange |
180 |
(Department of Futurology) |
181 |
- parameter lists to gmredi_?transport: added tracerIdentity |
182 |
- added new key indices to tamc.h |
183 |
o external_fields_load: |
184 |
- added this routine to TAF list |
185 |
- needed to make some common blocks global and additional storing |
186 |
along the same lines as exf package (checkpoint_lev?_directives.h) |
187 |
o exf: |
188 |
updated external forcing package |
189 |
- bug fixes carried over from ecco-branch |
190 |
(missing OBCS_OPTIONS.h in two routines) |
191 |
- enable easy to use "no forcing". |
192 |
- added exf I/O for atmospheric loading |
193 |
- added exf I/O for runoff data |
194 |
- transfered scaling between exf <-> MITgcm to exf namelist |
195 |
o Adding new verification experiment global_with_exf. |
196 |
This verification experiment is almost identical to the |
197 |
global_ocean.90x40x15 experiment, except that it uses |
198 |
the exf (and cal) package instead of the |
199 |
external_fields_load routine (cf README file). |
200 |
|
201 |
checkpoint46m_post |
202 |
o modified testscript to return -- when there are no lines of "%MON" output |
203 |
o added new flag "quasiHydrostatic" for QH mode |
204 |
and new routine quasihydrostaticterms() |
205 |
o added 2*Omega*Cos(phi)*W to u equations (non-hydrostatic Coriolis term) |
206 |
New routine: mom_u_coriolis_nh() |
207 |
o fixed sign of non-hydrostatic metric terms in u and v equations. |
208 |
Depend on correct sign of gravitySign. |
209 |
o added work.pc* to .cvsignore in aim directories using ifc |
210 |
o introduced flag for controllin non-hydrostatic metric terms |
211 |
o removed gravitySign from PARM04 |
212 |
o switched to g77/ifc from pgf77 in aim experiments |
213 |
|
214 |
checkpoint46l_post |
215 |
o made convective adjustment work with pressure coordinates: |
216 |
- changed the direction of k-loop in convective_adjustment.F for the |
217 |
case of pressure coordinates (OCEANICP,ATMOSPHERIC buoyancyRelation) |
218 |
- adjusted the reference pressure k-index in convective_adjustment.F |
219 |
- adjusted the convection condition in convect.F (in analogy to |
220 |
calc_ivdc.F) |
221 |
- convective_adjustment no longer computes anything on the halos |
222 |
- removed the warnings about negative salinity from find_rho.F and |
223 |
find_alpha.F; instead the new routine look_for_neg_salinity, called |
224 |
at the beginning of find_rho, find_alpha, and find_beta, does a |
225 |
check of the entire slice, if CPP-option |
226 |
CHECK_SALINITY_FOR_NEGATIVE_VALUES is defined |
227 |
checkpoint46l_pre |
228 |
|
229 |
checkpoint46k_post |
230 |
o fixed the verification/global_ocean.90x40x15 experiment: |
231 |
- new bathymetry (the world according to A., JMC, and M.) |
232 |
- new initial fields and forcing fields (*.bin files) |
233 |
- new POLY3.COEFFS (for the next release one should switch to a full |
234 |
equation of state: JMD95P or MDJWF) |
235 |
- fixed several errors and redundancies in the data file |
236 |
- experiment uses looped cells |
237 |
- added matlab directory with diagnostic scripts for plotting of output |
238 |
|
239 |
o S/R aim_initialise.F replace S/R aim_init from file aim_do_inphys.F: |
240 |
- read AIM physics parameters from a file (data.aimphys) |
241 |
- set defaults values = F.Molteni paper (Clim.Dyn., 2002) |
242 |
|
243 |
checkpoint46j_post |
244 |
o split calc_exact_eta in 2 S/R : integr_continuity & update_etaH |
245 |
o move wVel computation at the end of the time step, in S/R integr_continuity |
246 |
o create specific S/R to exchange T,S before DYNAMICS (for stagger time step) |
247 |
o update timeave pkg for wVel diagnostic. |
248 |
checkpoint46j_pre |
249 |
|
250 |
checkpoint46i_post |
251 |
o Clean up AIM package (and keep the results unchanged): |
252 |
include CPP_OPTION and use IMPLICT NONE in all routines ; |
253 |
declare all the variables _RL ; use _d 0 for all numerical constants. |
254 |
use ifdef ALLOW_AIM everywhere. And now AIM can be used with g77 ! |
255 |
|
256 |
checkpoint46h_post |
257 |
o cleaned up the use of rhoNil and rhoConst. |
258 |
- rhoNil should only appear in the LINEAR equation of state, everywhere |
259 |
else rhoNil is replaced by rhoConst, e.g. find_rho computes rho-rhoConst |
260 |
and the dynamical equations are all divided by rhoConst |
261 |
o introduced new parameter rhoConstFresh, a reference density of fresh |
262 |
water, to remove the fresh water flux's dependence on rhoNil. The default |
263 |
value is 999.8 kg/m^3 |
264 |
o cleanup up external_forcing.F and external_forcing_surf.F |
265 |
- can now be used by both OCEANIC and OCEANICP |
266 |
checkpoint46h_pre |
267 |
|
268 |
Added code to convert surface volume flux (fresh water) into |
269 |
a mass flux when using P coordinates in the ocean (OCEANICP). |
270 |
Note: It assumes you have set rho0=rhoConst=density of fresh water. |
271 |
|
272 |
checkpoint46g_post |
273 |
o Include a new diagnostic variable phiHydLow for the ocean model |
274 |
- in z-coordinates, it is the bottom pressure anomaly |
275 |
- in p-coordinates, it is the sea surface elevation |
276 |
- in both cases, these variable have global drift, reflecting the mass |
277 |
drift in z-coordinates and the volume drift in p-coordinates |
278 |
- included time averaging for phiHydLow, be aware of the drift! |
279 |
o depth-dependent computation of Bo_surf for pressure coordinates |
280 |
in the ocean (buoyancyRelation='OCEANICP') |
281 |
- requires a new routine (FIND_RHO_SCALAR) to compute density with only |
282 |
Theta, Salinity, and Pressure in the parameter list. This routine is |
283 |
presently contained in find_rho.F. This routine does not give the |
284 |
correct density for 'POLY3', which would be a z-dependent reference |
285 |
density. |
286 |
o cleaned up find_rho |
287 |
- removed obsolete 'eqn' from the parameter list. |
288 |
o added two new verification experiments: gop and goz |
289 |
(4x4 degree global ocean, 15 layers in pressure and height coordinates) |
290 |
checkpoint46g_pre |
291 |
|
292 |
checkpoint46f_post |
293 |
o Enable tangent linear (forward mode) gradient checks: |
294 |
- extended active file handling to g_... files |
295 |
- added TANGENT_SIMULATION to theSimulationMode |
296 |
- extended grdchk package accordingly |
297 |
o added tangent linear setup in verification/carbon/code_ftl |
298 |
o added adjoint setup in verification/front_relax/code_ad |
299 |
|
300 |
checkpoint46e_post |
301 |
o Added new equation of state -> MDJWF |
302 |
- EOS of McDougall et al., 2002, JAOT, submitted |
303 |
- caveat: the equation of state is only valid for a smaller (more |
304 |
realistic?) range of values than JMD95P/Z and UNESCO |
305 |
- added masks to the calculation of pressure in store_pressure |
306 |
- added more check values for density in check_eos (ini_eos.F), some of |
307 |
the old check values are out of the range of the MDJWF-eos, so don't |
308 |
expect perfect matches for those |
309 |
checkpoint46e_pre |
310 |
|
311 |
checkpoint46d_post |
312 |
o fixed store_pressure to work with both buoyancy relation = 'OCEANIC' and |
313 |
'OCEANICP', also initialised field pressure correctly in ini_eos in the |
314 |
case of pressure coordinates. eosType='JMD95Z' in |
315 |
combination with buoyancyRelation='OCEANICP' now causes an error. |
316 |
o Changed p = pressure(i,j,k,bi,bj) to p = pressure(i,j,kRef,bi,bj) |
317 |
in find_alpha/beta. |
318 |
checkpoint46d_pre |
319 |
|
320 |
o Changed p = pressure(i,j,k,bi,bj) to p = pressure(i,j,kRef,bi,bj) |
321 |
so that JMD95Z and JMD95P give approptiate static stability in find_rho. |
322 |
|
323 |
checkpoint46c_post |
324 |
o Changes necessary for ocean in p-coordinates |
325 |
- Added new buoyancy relation = 'OCEANICP' |
326 |
- Added new parameters = gravitySign (this used to be contained inside |
327 |
the factor dRdZ which I added when we first switched to R coordinates). |
328 |
X GM/Redi is not compatible (yet) |
329 |
X bottom drag and no-slip need to be debugged. |
330 |
checkpoint46c_pre |
331 |
|
332 |
checkpoint46b_post |
333 |
o Added new equation of state -> JMD95Z and JMD95P |
334 |
- EOS of Jackett and McDougall, 1995, JAOT |
335 |
- moved all EOS parameters into EOS.h |
336 |
- new routines ini_eos.F, store_pressure.F |
337 |
o Added UNESCO EOS, but not recommended because it requires |
338 |
in-situ temperature (see JMD95) |
339 |
o Modified formatting for knudsen2.f in utils/knudsen2 and added |
340 |
unesco.f to be used with POLY3 |
341 |
checkpoint46b_pre |
342 |
|
343 |
checkpoint46a_post |
344 |
o Added (atmospheric) pressure loading |
345 |
- new field in FFIELDS.h, etc... |
346 |
- new cpp flag ATMOSPHERIC_LOADING |
347 |
o Changed hFacC to _hFacC in calc_phi_hyd.F |
348 |
o Added SHORTWAVE_HEATING to some files for consistency |
349 |
checkpoint46a_pre |
350 |
|
351 |
checkpoint46 |
352 |
further merges from release1_p5 |
353 |
o new ctrl package |
354 |
- adopted from ECCO environment to enable optimization |
355 |
- added Eliassen Palm fluxes to controls |
356 |
o added Eliassen Palm flux controls to gradient check package |
357 |
o cost package |
358 |
- Compatible with new ctrl package |
359 |
- added Eliassen Palm cost hooks |
360 |
- modif's of existing cost functions |
361 |
o Adjoint-related bug fixes in kpp: |
362 |
- kpp_calc: sore of kpphbl avoids recomputation/call to S/R kppmix |
363 |
- kpp_routines: store of Rib avoids partial recomputation bug of TAF. |
364 |
o autodiff package |
365 |
- flow directives: |
366 |
+ new for S/R do_the_model_io |
367 |
+ enhanced for S/R checkpoint |
368 |
+ new gmredi.flow |
369 |
- Introduced CPP option ALLOW_AUTODIFF_MONITOR to |
370 |
disable adcommon.h/g_common.h by default. |
371 |
- fixed problem for adjoint of global_max |
372 |
o modified verification/carbon/ |
373 |
o New setup for adjoint of global_ocean |
374 |
o added hooks for Eliassen Palm fluxes (dynamics.F, ini_fields.F) |
375 |
o cleaned TAF keys iikey, idkey (dynamics.F, thermodynamics.F) |
376 |
o incorporated hooks for sponge layer code (external_forcing.F) |
377 |
|
378 |
checkpoint45d_post |
379 |
o import 2 fixes from release1_p5 (CG2D.h & packages_readparms.F) |
380 |
o use recip_dx*,recip_dy* instead of /dx*,/dy* in orlanski_E,W,N,S |
381 |
(affects plume_on_slope "cg2d" output) |
382 |
o add new parameter deltaTfreesurf for free-surface time stepping |
383 |
o exchange gT,gS when using staggerTimeStep & GM & Oly<4 |
384 |
o NONLIN_FRSURF: rescale gNm1 to get a better conservation with A-B |
385 |
o change phiMin in exp1 (to agree with documentation) |
386 |
o add new flags: |
387 |
- T,S forcing inside/outside Adams-Bashforth |
388 |
- turn off temp,salt advection and/or forcing |
389 |
- for each tracer, define internal flag for multiDimAdvection & A-B |
390 |
o monitor: change definition of KE to check conservation of Energy |
391 |
|
392 |
checkpoint45c_post |
393 |
o Removed f90 comments ("!") |
394 |
o Modified initialisations to enable exf package for MITgcm |
395 |
(without invoking ECCO_PACKAGE). |
396 |
o modifications to gradient check package (Martin Losch) |
397 |
- enable centered differences |
398 |
- modified format of standard output |
399 |
|
400 |
checkpoint45b_post |
401 |
o cleaned exchange calls in cg2d |
402 |
o Included CPP option ALLOW_AUTODIFF_MONITOR in forward_step |
403 |
o included CPP option SINGLE_LAYER_MODE in thermodynamics |
404 |
to configure barotropic setup (Martin Losch) |
405 |
o moved some initialisations within bi,bj loop in dynamics |
406 |
(TAF flow dependency) |
407 |
o in initialise_varia, moved call packages_init_variables |
408 |
before convective_adjustment_init |
409 |
(corresponding adjustment of TAF tape initialisation). |
410 |
o in calc_diffusivity/calc_viscosity extended domain to full overlap. |
411 |
|
412 |
checkpoint45a_post |
413 |
o compute Rho over the whole domain (extend iMin,jMin indices) |
414 |
o specific exchange for X-slice domain (case Ny=1) |
415 |
o argument futureIter of S/R obcs_calc missing in exp4 & internal_wave /code |
416 |
o solve_for_pressure : set cg2d_x to zero where OBCS are applied |
417 |
|
418 |
checkpoint44h_post = checkpoint45 |
419 |
o Fix to restore differentiability: |
420 |
place at which packages_init_variables is called in |
421 |
initialise_varia is crucial for reverse control flow. |
422 |
o modifications in GMRedi to enable stable adjoint |
423 |
(corresponding modif's in thermodynamics, dynamics) |
424 |
o added missing hook for sponge layer code in external_forcing |
425 |
o modified test cost functions |
426 |
o storing in gad_advection (plus removal of write(0,*)! |
427 |
o missing headers TR1.h added in convective_adjustment routines |
428 |
o errorMessageUnit set non-equal zero in eeboot |
429 |
(conflict on some platforms, e.g. IBM SP3) |
430 |
o modified carbon verif. |
431 |
|
432 |
checkpoint44h_pre |
433 |
o vertical grid option: allow to put the Interface at the middle between |
434 |
2 cell-centers ; replace delZ in pkg/kpp by drF. |
435 |
o GM Advective form: Tracers are advected using the residual transport (= |
436 |
Euler+GM-bolus); set param GM_AdvSeparate=T to return to previous form. |
437 |
|
438 |
checkpoint44g_post |
439 |
o fix surface correction term in multiDimAdvection (affects plume_on_slope |
440 |
results) ; ifdef missing in thermodynamics.F |
441 |
|
442 |
checkpoint44f_pre,checkpoint44f_post |
443 |
o added PTRACERS package |
444 |
This allows an arbitrary number of passive tracers to be integrated |
445 |
forward simultaneously with the dynamicaly model. |
446 |
+ Implemented so far: |
447 |
- basic forward algorithm (time-stepping, advection, diffusion, convection) |
448 |
- I/O and checkpointing |
449 |
- GM/Redi *but* using the GM/Redi coefficient of Salt |
450 |
+ Not implemented so far: |
451 |
- KPP |
452 |
- OBCS |
453 |
+ No specific example supplied (yet) but global_ocean.90x40x15 has the |
454 |
necessary data.ptracer file. Simply use -enable=ptracers and uncomment |
455 |
line in data.pkg. |
456 |
+ This package is disabled by default since it increases storage. |
457 |
|
458 |
|
459 |
o testscript: added -cleanup option |
460 |
o Shap_filt: a) new shap_filt S/R to use no-slip BC with S2 filter |
461 |
b) enable to filter 2D fields. |
462 |
|
463 |
checkpoint44e_pre,post |
464 |
o re-wrote convective adjustment to use pre-calculated weights |
465 |
Affects: model/src/convective_adjustment.F, convective_adjustment_ini.F |
466 |
Adds: model/src/convective_weights.F, convectively_mixtracer.F |
467 |
Original calls to convect() still in place. |
468 |
|
469 |
chkpt44d_pre,post |
470 |
o added missing EXCLUDE_MONITOR flags |
471 |
o changed "e" to "_d" in gmredi_slope_limit, gmredi_slope_psi |
472 |
(incompatible typ in MIN/MAX expressions caused problems |
473 |
on IBM SP3) |
474 |
o in genmake added variable MAKEDEPEND |
475 |
plus resetting for case SunOS |
476 |
o added timer_stats.c routine for IBM SP3 |
477 |
o removed variables in dynamics |
478 |
!!! I forgot to create tag chkpt44d_pre !!! |
479 |
!!! But chkpt44d_pre = chkpt44c_post !!! |
480 |
|
481 |
o real fresh water flux implemented with non-linear free-surface. |
482 |
|
483 |
chkpt44c_pre,post |
484 |
o few fix (mask in shap_s2, EmPmR in external_field_load, |
485 |
USE_NATURAL_BCS in solve_for_P); |
486 |
o add arguments myIter & myTime to S/R obcs_calc & solve_for_P |
487 |
|
488 |
checkpoint44b_pre/_post |
489 |
o merge of relevant stuff from the ecco-branch: |
490 |
- genmake: removed $S64 overwrite for case SunOS |
491 |
- pkg/exf: update and corrections for field swapping and obcs |
492 |
- pkg/ecco: parameter lists for the_model_main, the_main_loop |
493 |
harmonized between ECCO and MITgcm |
494 |
- pkg/autodiff: added flow directives for obcs, mdsio_gl_slice |
495 |
updated checkpointing_lev... lists for obcs |
496 |
- model/src: minor changes in forward_step, plot_field |
497 |
added directive for divided adjoint in the_main_loop |
498 |
- pkg/mdsio: added mdsio_gl_slice |
499 |
|
500 |
o check parameters & config (chkpt44a_pre,post) |
501 |
o OBC and NonLin_FrSurf. |
502 |
|
503 |
checkpoint44 |
504 |
o fix bug in mom_vi_del2uv |
505 |
o select when filters are applied ; add options to zonal_filter (data.zonfilt) |
506 |
o gmredi: fix Pb in the adiabatic form ; add options (.e.g. Bolus advection) |
507 |
o update AIM experiments (NCEP input files) |
508 |
o improve and extend diagnostics (Monitor, TimeAve with NonLin-FrSurf) |
509 |
o added some stuff for AD |
510 |
|
511 |
checkpoint43 |
512 |
o added two new advection tests |
513 |
o added global 4x4 degree ocean test |
514 |
o added test of H&S94 with free surface on cube |
515 |
o some final mods for gradient check, adjoint,... |
516 |
|
517 |
checkpoint42 |
518 |
Fixed AD-related problems: |
519 |
o Store directives up-to-date with re-arranged Adams-Bashforth |
520 |
(mainly thermodynamics.F) |
521 |
o New store directives for multi-dim. advection schemes |
522 |
* new CPP flag ALLOW_MULTI_DIM_ADVECTION |
523 |
* new common block and key passkey |
524 |
(mainly gad_advection.F) |
525 |
o Modified store directives for split of dynamics/thermodynamics |
526 |
for the case ALLOW_KPP |
527 |
o Cleaned argument list for timestep_tracer.F |
528 |
|
529 |
checkpoint41 |
530 |
o re-formatted/added comments in prologues of key subroutines |
531 |
to work with protex |
532 |
|
533 |
checkpoint40 |
534 |
o moved calc_gs, calc_gt and calc_tr1 back to model/src |
535 |
o added FLT (floats) package from Arne along with demo. experiment |
536 |
o re-arranged Adams-Bashforth scheme for tracers to allow easier |
537 |
implementation of forward-in-time forcing terms |
538 |
o more mods for non-linear free-surface including new variable |
539 |
o modified multi-dim method to work on cube |
540 |
o bug fix in KPP (from DM) |
541 |
|
542 |
checkpoint40pre9 |
543 |
o added DST advection schemes |
544 |
o fix MPI bug on SGI's |
545 |
o modifications for non-linear free-surface |
546 |
o fixed conditionals for fixed flow/passive tracer only mode |
547 |
o added tension/strain form of stress tensor |
548 |
o changes to commenting of arguments for TAF/TAMC |
549 |
o added/updated many matlab scripts |
550 |
o added multi-dimensional advection method for tracers |
551 |
|
552 |
checkpoint40pre8 |
553 |
o new package rw/ |
554 |
Added method for dumping intermediate local arrays: |
555 |
mdsio_writetile - same as mdsio_writefield except works from inside |
556 |
bi,bj loop |
557 |
mdsio_writelocal - same as mdsio_writetile except works for local |
558 |
arrays |
559 |
write_local_r? - higher-level wrapper for mdsio_writelocal |
560 |
Controlled by diagFreq. Defaults to zero (ie. no dumps) |
561 |
Example given at end of mom_vecinv.F that dumps some local arrays. |
562 |
o replaced calc_mom_rhs.F by mom_fluxform.F/mom_vecinc.F to |
563 |
control flux-form versus vector invariant code. |
564 |
Switch is runtime flag vectorInvariantMomentum (defaults to false) |
565 |
or genmake -DISABLE ... |
566 |
o TAMC-related update to cope with mom_fluxform.F/mom_vecinc.F |
567 |
(additional storing is now needed for case vecinv; |
568 |
currently nothing implemented to switch this storing off |
569 |
in case fluxform is used since no CPP flag available). |
570 |
|
571 |
checkpoint40pre7 |
572 |
o bug fix in dynamics.F |
573 |
calc_viscosity called before k=1,Nr loop. |
574 |
o added some #ifdef's ALLOW_PASSIVE_TRACER |
575 |
o cleaned initialisations in ctrl_map_ini |
576 |
related to TAF |
577 |
|
578 |
checkpoint40pre6 |
579 |
o split dynamics into thermodynamics & dynamics |
580 |
o corresponding split of calc_diffusivity into |
581 |
calc_diffusivity, calc_viscosity |
582 |
(kpp_calc_diff, kpp_calc_visc). |
583 |
o Added diffkr, kapgm to control variables |
584 |
o bug fix in gmredi_slope_limit for ldd97 |
585 |
|
586 |
checkpoint40pre5 |
587 |
o proper initialization of Adams-Bashforth time stepping |
588 |
+ uses forward step for very first step of model (iter=0) |
589 |
o re-generated *all* output since almost all output is modified by this |
590 |
change (sorry but we've been putting this one off for about 2 years) |
591 |
+ natl_box, aim.5l_Equatorial_Channel and aim.5l_zon-ave were |
592 |
*not* regenerated since they are already out of date and I don't |
593 |
want to create the impression that they are suddenly working. |
594 |
o removed KAP stuff for DEC's |
595 |
o modified/added many .cvsignore files in */code and */input |
596 |
o added new expt "solid-body.cs-32x32x1" |
597 |
+ this uses an alternative form of ini_curvilinear_grid.F which |
598 |
is not necessarily any better than the current one... |
599 |
|
600 |
checkpoint40pre4 |
601 |
o added re-initialisations of utrans,vtrans,rtrans in |
602 |
calc_common_factors which are crucial for TAF. |
603 |
o extended iMin,jMin range for calc_common_factors, |
604 |
calc_diffusivity |
605 |
o added option ALLOW_PASSIVE_TRACER |
606 |
o a few minor updates and fixes |
607 |
|
608 |
checkpoint40pre3 |
609 |
o created calc_gtr1.F for gad package |
610 |
o split the_main_loop and re-introduced forward_step |
611 |
o some recomputations of nIter0 for TAMC |
612 |
|
613 |
checkpoint40pre2 |
614 |
o Updated to latest branch of AJA, JMC |
615 |
o Added gradient check package |
616 |
o Added routines for passive tracer handling |
617 |
o Added carbon sequestration sensitivity experiment |
618 |
|
619 |
checkpoint39 |
620 |
o Modifications and fixes to support TAMC |
621 |
(mainly missing or wrong directives) |
622 |
o Corrected common blocks when fooling TAMC to produce adjoint I/O |
623 |
in addummy_in_stepping |
624 |
o Updated verification/vero/ (bug fixes in code/, correct pickups in input/) |
625 |
o Enabled "sliced" (xz-,yz-) exchanges, needed for obcs. |
626 |
o Included following packages: |
627 |
cal: calendar |
628 |
exf: external forcing |
629 |
ecco: essential routines to control use of cal, exf |
630 |
(NB: this is not the full ECCO environment) |
631 |
o disabled some packages by default in genmake |
632 |
|
633 |
checkpoint38 |
634 |
o see notes_c37_adj.txt |
635 |
o corrected wrong layer index in external_forcing_t |
636 |
o replaced the_main_loop/forward_step by the_main_loop |
637 |
o removed forceoutput from write_state |
638 |
(replaced criteria for initial/final state) |
639 |
o added runtime flag for pre-C35 pickups: usePickupBeforeC35 |
640 |
|
641 |
checkpoint37 |
642 |
o mdsio.F routine split and packaged in pkg/mdsio |
643 |
(mdsio.F and mdsio_gl.F still exist in eesupp/src but are not |
644 |
used unless genmake -disable=mdsio) |
645 |
o allow a more precise surface-geopotential gradient for the atmosphere |
646 |
taking into account the change in space of Alpha (=1/rho) at the surface. |
647 |
o variable "etaN" (replace cg2d_x) is the surface R-anomaly [r unit] |
648 |
(ocean: sea surface height [m] / atmos: surface pressure anomaly [Pa]) |
649 |
o Change units: Phi_Hydrostatic (=phiHyd), Phi_Surface(=cg2d_x) and |
650 |
Phi_NonHydrostatic (=cg3d_x), all have now the dimension of a potential |
651 |
[m2/s2] (pressure/rhoConst in the ocean ; Geo-Potential in the atmosphere). |
652 |
o Time-Average diagnostics defined as a pseudo package "timeave". |
653 |
o pickup files : old pickup (before c35) did not work with new code (c35 |
654 |
and after) unless activating 2 commented lines in initialise-varia. |
655 |
o Non-Hydrostatic pickup file Pb fixed. |
656 |
o bug with Rigid-lid version (since c35) fixed. |
657 |
|
658 |
checkpoint36 |
659 |
o implement Crank-Nickelson barotropic time-stepping |
660 |
o recover Time-Average diagnostics |
661 |
o fix Pb in checkpoint (PR#19) |
662 |
o AIM become a standard package ; clean unused CPP_kees |
663 |
|
664 |
checkpoint35 |
665 |
o subtantial rewrite of dynamics.f |
666 |
o allows staggered time-stepping |
667 |
o packaged obcs, aim (Atmospheric Physics), zonal_filt, shap_filt |
668 |
o added bottom drag (linear + quadratic) |
669 |
|
670 |
checkpoint34 |
671 |
o Modified exchange routines for correct reverse simulation. |
672 |
o Added verification output to comply with bug fix in impldiff.F |
673 |
as of c33 (exp2, natl_box). |
674 |
o Corrected store directives in dynamics.F |
675 |
o Corrected array boundaries in impldiff.F |
676 |
o Corrected array dimensions in gmredi. |
677 |
Added temporary scalars to avoid storage by TAMC (gmredi, kpp). |
678 |
o Added routine GLOBAL_SUM_INT to global_sum.F |
679 |
o Added mdsio_gl.F for ECCO purposes. |
680 |
|
681 |
checkpoint33 |
682 |
o Fixed bug in lower BC in impldiff.F |
683 |
o Fixed confusion about units of forcing arrays in FFIELDS.h |
684 |
namely Fu,Fv,Qnet,Qsw,EmPmR: |
685 |
- Removed verification/natl_box/code/external_fields_scale.F |
686 |
(did not differ from that in model/src) |
687 |
- Changed units of fu,fv,Qnet,Qsw,EmPmR back to proper units |
688 |
(see FFIELDS.h for description) |
689 |
- Scale fu,fv,Qnet,Qsw,EmPmR when used in external_forcing_surf.F, |
690 |
kpp_calc.F and kpp_transport_t.F |
691 |
- Removed model/src/external_fields_scale.F and calls to it |
692 |
- verification/natl_box uses flux data with "atmospheric" sign so |
693 |
a special version of external_fields_load.F is used to |
694 |
change the data as it's read in. This way, the arrays |
695 |
have the right units and signs at all times tha a user could |
696 |
possibly use them. |
697 |
o Corrected genmake to delete a temporary file during configuration |
698 |
|
699 |
checkpoint32 |
700 |
o Scaling of forcing fields moved from external_forcing.F to |
701 |
external_fields_scale.F, called directly after loading fields. |
702 |
o Surface relaxation terms added to surface tendency fields in |
703 |
extra subroutine. |
704 |
o impldiff.F changed to avoid storing by TAMC of huge intermediate |
705 |
fields. |
706 |
o Changes in GMRedi to comply with packaging structure. |
707 |
o Modifications and bug fixes in KPP package. |
708 |
o Verification experiment for KPP added (verification/nat_box/) |
709 |
|
710 |
checkpoint31 |
711 |
o Packages interface brought in line with defined standard |
712 |
-> new routines packages_...F |
713 |
-> changes in initialise_fixed.F, initialise_varia.F |
714 |
-> enables initialisation interface for ECCO package |
715 |
o Rescaling of external forcing fields immediately after read |
716 |
(removed from external_forcing.F) |
717 |
-> new: surface_fields_scale.F |
718 |
-> affects ini_foorcing.F, external_fields_load.F |
719 |
o Computation of surface tendencies before tendency updates. |
720 |
-> new: external_forcing_surf.F |
721 |
o Change of variable names useGMRedi, useKPP, useECCO |
722 |
o Further changes in dynamics.F, impldiff.F for TAMC |
723 |
o Tested for experiments 0, 2, 4. |
724 |
- not yet tested for GMRedi. |
725 |
- not yet tested for KPP. |
726 |
|
727 |
checkpoint30 |
728 |
o Updates for OBCs |
729 |
o New experiment for test OBCs (internal_wave) |
730 |
o New output.txt in verification/exp4/results/ |
731 |
o Updates for KPP: calculation of viscosity corrected, new variables |
732 |
kpp_dumpFreq and kpp_taveFreq |
733 |
o CPP'd calls to system routines for TARGET_CRAY_VECTOR (P.H.) |
734 |
and entry for T90 in genmake |
735 |
|
736 |
checkpoint29 |
737 |
o packaged KPP and GM/Redi |
738 |
|
739 |
checkpoint28 |
740 |
o Corrected initialization of hFacMin for difference vertical coordinates |
741 |
o Modified calculation of partial cell thickness to more robust algorithm |
742 |
o config_summary.F: replaced write_1D_... by write_0D_... |
743 |
read_write.F: added subroutines write_0D_... |
744 |
o routines calc_common_factors, calc_gs, calc_gt, calc_phi_hyd, find_rho: |
745 |
included some initialisations required for TAMC. |
746 |
o routine calc_isoslopes: modified sqrt computation to make it |
747 |
differentiable for zero argument. |
748 |
o routines dynamics, impldiff: |
749 |
included store directives and tape key computations required for TAMC. |
750 |
o ALL TAMC related changes are between ifdef ALLOW_AUTODIFF_TAMC |
751 |
|
752 |
checkpoint27 |
753 |
o fixed calc_isoslopes() to so that cancellation of terms in tensor |
754 |
formulation works properly with variable resolution. |
755 |
o restructured the calling tree between the_model_main() and dynamics() |
756 |
o split initialise() into initialise_fixed() and initialise_varia() |
757 |
o introduced initialization file pSurfInitFile for free surface (cg2d_x) |
758 |
in new routine ini_psurf() |
759 |
|
760 |
checkpoint26 |
761 |
o moved some #ifdef in optional routines to encompass everything |
762 |
except the argument declarations. This was using up some memory. |
763 |
o fixed macros like _ADT to work with both versions of Linux cpp |
764 |
o cleaned up some unused and uninitialized variables |
765 |
(helps when debugging with strict compile options) |
766 |
o split up CPP_EEOPTIONS.h into options and macros |
767 |
(macros are now in eesupp/inc/CPP_EEMACROS.h) |
768 |
o patch for 2 processor JAM mode |
769 |
|
770 |
checkpoint25 |
771 |
o updates for OBCs and NH |
772 |
o dramatically reduced memory usage |
773 |
o added "JAM" routines for use on Artic network |
774 |
o parameterization of convection by implicit vertical diffusion |
775 |
o added a diagnostic of convective index |
776 |
o moved call to calc_divg_hat() from dynamics() to solve_for_pressure() |
777 |
(this is was partly for OBCs and NH but also in anticipation of |
778 |
an explicit free-surface option) |
779 |
o initial state input files for U and V |
780 |
|
781 |
checkpoint24 |
782 |
o Removed some spurious "REAL"s |
783 |
o Updated KPP to latest version from JPL |
784 |
o Added a COS(latitude) in horizontal viscosity |
785 |
|
786 |
checkpoint23 |
787 |
o Added del^4 diffusion for salt and temperature |
788 |
o Multiple minor fixes (implicit none's, arguments to global_sum, etc.) |
789 |
|
790 |
checkpoint22 |
791 |
o Made many small changes: implicit none, format(i5), etc... |
792 |
o Introduced KPP mixing scheme (compiles and runs but *not* tested) |
793 |
|
794 |
checkpoint21 |
795 |
o Introduced new I/O package that supports direct-access records (mdsio.F) |
796 |
o Split-up mid-level I/O routines |
797 |
o Updated input data for experiments 0,1,2,4 and 5 |
798 |
o Finally removed the genmake.* scripts (everyone seems happy with genmake) |
799 |
|
800 |
checkpoint20 |
801 |
o Tidied up open-boundaries some more |
802 |
o Added non-hydrostatic code |
803 |
o Fixed some input data |
804 |
o Modified MPI start-up to work in "coupled" context. |
805 |
|
806 |
checkpoint19 |
807 |
o Tidied up directory structure for verification experiments. |
808 |
o Added new code for open boundaries. |
809 |
o Corrected advective terms near boundaries. |
810 |
o Added CPP flag for "old UV" geometry and corrected spherical geometry. |
811 |
o Implemented free-slip/no-slip BCs and del^4 dissipation. |
812 |
o Split default parameters out of ini_parms.F into set_defaults.F. |
813 |
|
814 |
checkpoint18 |
815 |
o Further changes for optional code generation with |
816 |
CPP - ugh! |
817 |
|
818 |
checkpoint17 |
819 |
o Beginning to incorporating modularity |
820 |
for supporting atmos. and ocean configs. |
821 |
o Small changes for TAMC complinace |
822 |
o New ultra simple test case. Barotropic box |
823 |
configuration "exp0". |
824 |
|
825 |
checkpoint16 |
826 |
o Minor changes to exchange routines to support |
827 |
accumulation in reverse mode. Adjoint of an |
828 |
assignment is an addition. |
829 |
o Changes to support compilation by TAMC (LR(1) |
830 |
syntax) and by g77 under Linux ( 72 character |
831 |
limit ). |
832 |
|
833 |
checkpoint15 |
834 |
Rewrote exchange routines to allow |
835 |
o General tile <-> tile connectivity. |
836 |
o TAMC reverse mode flag ( is this really needed? ) |
837 |
o Variable width overlap and exchange region widths. |
838 |
o Hooks for sharing data via DMA, Arctic, Memory Channel, |
839 |
shmput/shmget, VIA, SYSV shmem and every other cool |
840 |
communication method. |
841 |
|
842 |
checkpoint14 |
843 |
Consistent isomorphism chages made. These allow dynamical core |
844 |
to support both pressure and meters as a vertical coordinate. |
845 |
|
846 |
checkpoint13 |
847 |
Release which corrects global_max problem and a couple |
848 |
of KGM time-averaging diagnostic errors. Also contains |
849 |
a double-gyre experiment setup as well as the baseline |
850 |
4 degree global template. The double gyre template is |
851 |
organised to allow it to be applied automatically to |
852 |
the baseline configuration. |
853 |
|
854 |
branch-point-rdot |
855 |
A few tidy-ups have been made bt the real point of this |
856 |
check-point (excuse the pun) is to allow a branch to be |
857 |
made for the purposes of adding the rDot alterations. |
858 |
I don't want to commit the rDot to the main-trunk until |
859 |
we've actually tested it... |
860 |
|
861 |
checkpoint12 |
862 |
Tidy up and added generic genmake script (formerly |
863 |
configure script). |
864 |
Released to testers. |
865 |
|
866 |
checkpoint11 |
867 |
Re-arranged initialise() and introduced ini_vertical_grid() |
868 |
and ini_masks_etc(). |
869 |
|
870 |
checkpoint10 |
871 |
Introduced the configure script. |
872 |
Separated out the diags package. |
873 |
Added template for writing time-averages of tile/intermediate quantities. |
874 |
Updated SIZE.h in exp2 and made the data big-endian. |
875 |
|
876 |
checkpoint9 |
877 |
MPI fix. Filename changes (meta/data). salbin*y stuff. |
878 |
General tidy-up. |
879 |
SST.bin SSS.bin added to verification/exp2 |
880 |
|
881 |
checkpoint8 |
882 |
Inludes time-dependant forcing and time-averaging. |
883 |
Works at 4x4 and 2.8125x2.8125 global |
884 |
|
885 |
checkpoint7-4degree-ref |
886 |
Branch from checkpoint7. Released for application to |
887 |
global modeling and general modeling - process studies, regional |
888 |
model etc.. |
889 |
Configured for verification/exp2 - 4 degree global, two process. |
890 |
|
891 |
checkpoint7 |
892 |
Created by cnh |
893 |
Simple 4 degree global simulation |
894 |
configuration. |
895 |
Includes climatological wind, |
896 |
hydrography, bathymetry etc |
897 |
Does not include Heat Flux and E-P |
898 |
|
899 |
checkpoint6 |
900 |
Created by cnh |
901 |
merged checkpoint5 with CD and |
902 |
spherical polar code. |
903 |
Still some problems with GM Redi |
904 |
abd full topography. |
905 |
|
906 |
checkpoint5 |
907 |
Created by AJA |
908 |
contains updates to GM Redi |
909 |
and reordered timestepping loop |
910 |
for Implicit Diffusion. |
911 |
|
912 |
checkpoint4 |
913 |
Created by cnh |
914 |
Contains memory saving macros ( _rdxc etc..) for when |
915 |
grid variations are limited. |
916 |
Contains polynomial coeffs for EOS and knudsen program |
917 |
for generation. |
918 |
|
919 |
checkpoint3 |
920 |
Created by cnh |
921 |
Added extra flags for runtime options |
922 |
Added this file |
923 |
Added implicit free-surface |
924 |
Added pre-conditioner |
925 |
Put compare01 into repository ( compare01/... ) |
926 |
Put data and eedata into repository ( verification/exp1 ) |
927 |
|
928 |
checkpoint2 |
929 |
Created by AJA |
930 |
Added GM/Redi |
931 |
|
932 |
checkpoint1 |
933 |
Created by AJA |
934 |
Restructured kloop in dynamics.F |
935 |
Added non-linear EOS |
936 |
|
937 |
baseline |
938 |
Original checked in code |