3 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
4 |
============================== |
============================== |
5 |
|
|
6 |
|
o new pkg: aim_v23 |
7 |
|
= F. Molteni atmos.physics (SPEEDY, ver23) adapted to MITgcm. |
8 |
|
for now, keep the same surface forcing as before. |
9 |
|
- Part-Cell implemented into AIM; check that Heat & Water are conserved |
10 |
|
- aim.5l_cs: use new aim pkg (run 10yr & get better results than before) |
11 |
|
- aim.5l_LatLon & aim.5l_Equatorial_Channel are still using old aim pkg |
12 |
|
o add diagnostic of surface correction term in monitor output |
13 |
|
|
14 |
|
checkpoint47 |
15 |
|
differentiable version of checkpoint46n_post |
16 |
|
o bug fix in quasihydrostaticterms |
17 |
|
o * "clean" adjoint code (in terms of extensive recomputations) |
18 |
|
can now be obtained for all GMREDI options (i.e. for |
19 |
|
- GM_VISBECK_VARIABLE_K |
20 |
|
- GM_NON_UNITY_DIAGONAL |
21 |
|
- GM_EXTRA_DIAGONAL |
22 |
|
- GM_BOLUS_ADVEC ) |
23 |
|
* However, wrong gradient check problem remains unsolved. |
24 |
|
* New CPP options have been introduced for different |
25 |
|
tapering schemes |
26 |
|
o external_fields_load now part of differentiation list |
27 |
|
o pressure needs multiple storing; |
28 |
|
would be nice to have store_pressure at beginning or |
29 |
|
end of forward_step, e.g. by having phiHyd global (5-dim.) |
30 |
|
(NB: pressure is needed for certain cases in find_rho, |
31 |
|
which is also invoked through convective_adjustment). |
32 |
|
o recomputations in find_rho for cases |
33 |
|
'JMD95'/'UNESCO' or 'MDJWF' are OK. |
34 |
|
o #define ATMOSPHERIC_LOADING should be differentiable |
35 |
|
o ini_forcing shifted to begining of initialise_varia |
36 |
|
o Incorporating QNVS line search routines into MITgcm |
37 |
|
(this is separate code, and therefore not under pkg) |
38 |
|
lsopt/, optim/ |
39 |
|
o Updated verification/carbon/ |
40 |
|
- #define GM_VISBECK_VARIABLE_K |
41 |
|
- GM_taper_scheme = 'ldd97' |
42 |
|
|
43 |
|
checkpoint46n_post |
44 |
|
Merging from release1_p8: |
45 |
|
o verification/natl_box: |
46 |
|
updating new external_fields_load routine |
47 |
|
o New package: pkg/seaice |
48 |
|
Sea ice model by D. Menemenlis (JPL) and Jinlun Zhang (Seattle). |
49 |
|
The sea-ice code is based on Hibler (1979-1980). |
50 |
|
Two sea-ice dynamic solvers, ADI and LSR, are included. |
51 |
|
In addition to computing prognostic sea-ice variables and diagnosing |
52 |
|
the forcing/external data fields that drive the ocean model, |
53 |
|
SEAICE_MODEL also sets theta to the freezing point under sea-ice. |
54 |
|
The implied surface heat flux is then stored in variable |
55 |
|
surfaceTendencyTice, which is needed by KPP package (kpp_calc.F and |
56 |
|
kpp_transport_t.F) to diagnose surface buoyancy fluxes and for the |
57 |
|
non-local transport term. Because this call precedes model |
58 |
|
thermodynamics, temperature under sea-ice may not be "exactly" at |
59 |
|
the freezing point by the time theta is dumped or time-averaged. |
60 |
|
N.B.: fairly large differences in accuracy occur across |
61 |
|
different platforms/compilers; comparison between |
62 |
|
g77 (Linux) and F77 (SGI Origin 2000/MIPSpro) |
63 |
|
generated output gives: |
64 |
|
T S U V |
65 |
|
C D M c m s m s m s m s |
66 |
|
n p a R g m m e . m m e . m m e . m m e . |
67 |
|
f n k u 2 i a a d i a a d i a a d i a a d |
68 |
|
g d e n d n x n . n x n . n x n . n x n . |
69 |
|
|
70 |
|
Y Y Y Y 5 5 7 7 7 8 10 9 6 6 6 6 7 5 7 5 7 FAIL lab_sea |
71 |
|
o GAD: |
72 |
|
- generated new common blocks to account for call of |
73 |
|
same gad routines with differing traceridentities |
74 |
|
(needed to modify tracerIdentity indices in GAD.h) |
75 |
|
- generated separate common blocks for case useCubedSphereExchange |
76 |
|
(Department of Futurology) |
77 |
|
- parameter lists to gmredi_?transport: added tracerIdentity |
78 |
|
- added new key indices to tamc.h |
79 |
|
o external_fields_load: |
80 |
|
- added this routine to TAF list |
81 |
|
- needed to make some common blocks global and additional storing |
82 |
|
along the same lines as exf package (checkpoint_lev?_directives.h) |
83 |
|
o exf: |
84 |
|
updated external forcing package |
85 |
|
- bug fixes carried over from ecco-branch |
86 |
|
(missing OBCS_OPTIONS.h in two routines) |
87 |
|
- enable easy to use "no forcing". |
88 |
|
- added exf I/O for atmospheric loading |
89 |
|
- added exf I/O for runoff data |
90 |
|
- transfered scaling between exf <-> MITgcm to exf namelist |
91 |
|
o Adding new verification experiment global_with_exf. |
92 |
|
This verification experiment is almost identical to the |
93 |
|
global_ocean.90x40x15 experiment, except that it uses |
94 |
|
the exf (and cal) package instead of the |
95 |
|
external_fields_load routine (cf README file). |
96 |
|
|
97 |
|
checkpoint46m_post |
98 |
|
o modified testscript to return -- when there are no lines of "%MON" output |
99 |
|
o added new flag "quasiHydrostatic" for QH mode |
100 |
|
and new routine quasihydrostaticterms() |
101 |
|
o added 2*Omega*Cos(phi)*W to u equations (non-hydrostatic Coriolis term) |
102 |
|
New routine: mom_u_coriolis_nh() |
103 |
|
o fixed sign of non-hydrostatic metric terms in u and v equations. |
104 |
|
Depend on correct sign of gravitySign. |
105 |
|
o added work.pc* to .cvsignore in aim directories using ifc |
106 |
|
o introduced flag for controllin non-hydrostatic metric terms |
107 |
|
o removed gravitySign from PARM04 |
108 |
|
o switched to g77/ifc from pgf77 in aim experiments |
109 |
|
|
110 |
|
checkpoint46l_post |
111 |
|
o made convective adjustment work with pressure coordinates: |
112 |
|
- changed the direction of k-loop in convective_adjustment.F for the |
113 |
|
case of pressure coordinates (OCEANICP,ATMOSPHERIC buoyancyRelation) |
114 |
|
- adjusted the reference pressure k-index in convective_adjustment.F |
115 |
|
- adjusted the convection condition in convect.F (in analogy to |
116 |
|
calc_ivdc.F) |
117 |
|
- convective_adjustment no longer computes anything on the halos |
118 |
|
- removed the warnings about negative salinity from find_rho.F and |
119 |
|
find_alpha.F; instead the new routine look_for_neg_salinity, called |
120 |
|
at the beginning of find_rho, find_alpha, and find_beta, does a |
121 |
|
check of the entire slice, if CPP-option |
122 |
|
CHECK_SALINITY_FOR_NEGATIVE_VALUES is defined |
123 |
|
checkpoint46l_pre |
124 |
|
|
125 |
|
checkpoint46k_post |
126 |
|
o fixed the verification/global_ocean.90x40x15 experiment: |
127 |
|
- new bathymetry (the world according to A., JMC, and M.) |
128 |
|
- new initial fields and forcing fields (*.bin files) |
129 |
|
- new POLY3.COEFFS (for the next release one should switch to a full |
130 |
|
equation of state: JMD95P or MDJWF) |
131 |
|
- fixed several errors and redundancies in the data file |
132 |
|
- experiment uses looped cells |
133 |
|
- added matlab directory with diagnostic scripts for plotting of output |
134 |
|
|
135 |
|
o S/R aim_initialise.F replace S/R aim_init from file aim_do_inphys.F: |
136 |
|
- read AIM physics parameters from a file (data.aimphys) |
137 |
|
- set defaults values = F.Molteni paper (Clim.Dyn., 2002) |
138 |
|
|
139 |
|
checkpoint46j_post |
140 |
|
o split calc_exact_eta in 2 S/R : integr_continuity & update_etaH |
141 |
|
o move wVel computation at the end of the time step, in S/R integr_continuity |
142 |
|
o create specific S/R to exchange T,S before DYNAMICS (for stagger time step) |
143 |
|
o update timeave pkg for wVel diagnostic. |
144 |
|
checkpoint46j_pre |
145 |
|
|
146 |
|
checkpoint46i_post |
147 |
|
o Clean up AIM package (and keep the results unchanged): |
148 |
|
include CPP_OPTION and use IMPLICT NONE in all routines ; |
149 |
|
declare all the variables _RL ; use _d 0 for all numerical constants. |
150 |
|
use ifdef ALLOW_AIM everywhere. And now AIM can be used with g77 ! |
151 |
|
|
152 |
|
checkpoint46h_post |
153 |
|
o cleaned up the use of rhoNil and rhoConst. |
154 |
|
- rhoNil should only appear in the LINEAR equation of state, everywhere |
155 |
|
else rhoNil is replaced by rhoConst, e.g. find_rho computes rho-rhoConst |
156 |
|
and the dynamical equations are all divided by rhoConst |
157 |
|
o introduced new parameter rhoConstFresh, a reference density of fresh |
158 |
|
water, to remove the fresh water flux's dependence on rhoNil. The default |
159 |
|
value is 999.8 kg/m^3 |
160 |
|
o cleanup up external_forcing.F and external_forcing_surf.F |
161 |
|
- can now be used by both OCEANIC and OCEANICP |
162 |
|
checkpoint46h_pre |
163 |
|
|
164 |
|
Added code to convert surface volume flux (fresh water) into |
165 |
|
a mass flux when using P coordinates in the ocean (OCEANICP). |
166 |
|
Note: It assumes you have set rho0=rhoConst=density of fresh water. |
167 |
|
|
168 |
|
checkpoint46g_post |
169 |
|
o Include a new diagnostic variable phiHydLow for the ocean model |
170 |
|
- in z-coordinates, it is the bottom pressure anomaly |
171 |
|
- in p-coordinates, it is the sea surface elevation |
172 |
|
- in both cases, these variable have global drift, reflecting the mass |
173 |
|
drift in z-coordinates and the volume drift in p-coordinates |
174 |
|
- included time averaging for phiHydLow, be aware of the drift! |
175 |
|
o depth-dependent computation of Bo_surf for pressure coordinates |
176 |
|
in the ocean (buoyancyRelation='OCEANICP') |
177 |
|
- requires a new routine (FIND_RHO_SCALAR) to compute density with only |
178 |
|
Theta, Salinity, and Pressure in the parameter list. This routine is |
179 |
|
presently contained in find_rho.F. This routine does not give the |
180 |
|
correct density for 'POLY3', which would be a z-dependent reference |
181 |
|
density. |
182 |
|
o cleaned up find_rho |
183 |
|
- removed obsolete 'eqn' from the parameter list. |
184 |
|
o added two new verification experiments: gop and goz |
185 |
|
(4x4 degree global ocean, 15 layers in pressure and height coordinates) |
186 |
|
checkpoint46g_pre |
187 |
|
|
188 |
|
checkpoint46f_post |
189 |
|
o Enable tangent linear (forward mode) gradient checks: |
190 |
|
- extended active file handling to g_... files |
191 |
|
- added TANGENT_SIMULATION to theSimulationMode |
192 |
|
- extended grdchk package accordingly |
193 |
|
o added tangent linear setup in verification/carbon/code_ftl |
194 |
|
o added adjoint setup in verification/front_relax/code_ad |
195 |
|
|
196 |
|
checkpoint46e_post |
197 |
|
o Added new equation of state -> MDJWF |
198 |
|
- EOS of McDougall et al., 2002, JAOT, submitted |
199 |
|
- caveat: the equation of state is only valid for a smaller (more |
200 |
|
realistic?) range of values than JMD95P/Z and UNESCO |
201 |
|
- added masks to the calculation of pressure in store_pressure |
202 |
|
- added more check values for density in check_eos (ini_eos.F), some of |
203 |
|
the old check values are out of the range of the MDJWF-eos, so don't |
204 |
|
expect perfect matches for those |
205 |
|
checkpoint46e_pre |
206 |
|
|
207 |
|
checkpoint46d_post |
208 |
|
o fixed store_pressure to work with both buoyancy relation = 'OCEANIC' and |
209 |
|
'OCEANICP', also initialised field pressure correctly in ini_eos in the |
210 |
|
case of pressure coordinates. eosType='JMD95Z' in |
211 |
|
combination with buoyancyRelation='OCEANICP' now causes an error. |
212 |
|
o Changed p = pressure(i,j,k,bi,bj) to p = pressure(i,j,kRef,bi,bj) |
213 |
|
in find_alpha/beta. |
214 |
|
checkpoint46d_pre |
215 |
|
|
216 |
|
o Changed p = pressure(i,j,k,bi,bj) to p = pressure(i,j,kRef,bi,bj) |
217 |
|
so that JMD95Z and JMD95P give approptiate static stability in find_rho. |
218 |
|
|
219 |
|
checkpoint46c_post |
220 |
|
o Changes necessary for ocean in p-coordinates |
221 |
|
- Added new buoyancy relation = 'OCEANICP' |
222 |
|
- Added new parameters = gravitySign (this used to be contained inside |
223 |
|
the factor dRdZ which I added when we first switched to R coordinates). |
224 |
|
X GM/Redi is not compatible (yet) |
225 |
|
X bottom drag and no-slip need to be debugged. |
226 |
|
checkpoint46c_pre |
227 |
|
|
228 |
|
checkpoint46b_post |
229 |
|
o Added new equation of state -> JMD95Z and JMD95P |
230 |
|
- EOS of Jackett and McDougall, 1995, JAOT |
231 |
|
- moved all EOS parameters into EOS.h |
232 |
|
- new routines ini_eos.F, store_pressure.F |
233 |
|
o Added UNESCO EOS, but not recommended because it requires |
234 |
|
in-situ temperature (see JMD95) |
235 |
|
o Modified formatting for knudsen2.f in utils/knudsen2 and added |
236 |
|
unesco.f to be used with POLY3 |
237 |
|
checkpoint46b_pre |
238 |
|
|
239 |
|
checkpoint46a_post |
240 |
|
o Added (atmospheric) pressure loading |
241 |
|
- new field in FFIELDS.h, etc... |
242 |
|
- new cpp flag ATMOSPHERIC_LOADING |
243 |
|
o Changed hFacC to _hFacC in calc_phi_hyd.F |
244 |
|
o Added SHORTWAVE_HEATING to some files for consistency |
245 |
|
checkpoint46a_pre |
246 |
|
|
247 |
|
checkpoint46 |
248 |
|
further merges from release1_p5 |
249 |
|
o new ctrl package |
250 |
|
- adopted from ECCO environment to enable optimization |
251 |
|
- added Eliassen Palm fluxes to controls |
252 |
|
o added Eliassen Palm flux controls to gradient check package |
253 |
|
o cost package |
254 |
|
- Compatible with new ctrl package |
255 |
|
- added Eliassen Palm cost hooks |
256 |
|
- modif's of existing cost functions |
257 |
|
o Adjoint-related bug fixes in kpp: |
258 |
|
- kpp_calc: sore of kpphbl avoids recomputation/call to S/R kppmix |
259 |
|
- kpp_routines: store of Rib avoids partial recomputation bug of TAF. |
260 |
|
o autodiff package |
261 |
|
- flow directives: |
262 |
|
+ new for S/R do_the_model_io |
263 |
|
+ enhanced for S/R checkpoint |
264 |
|
+ new gmredi.flow |
265 |
|
- Introduced CPP option ALLOW_AUTODIFF_MONITOR to |
266 |
|
disable adcommon.h/g_common.h by default. |
267 |
|
- fixed problem for adjoint of global_max |
268 |
|
o modified verification/carbon/ |
269 |
|
o New setup for adjoint of global_ocean |
270 |
|
o added hooks for Eliassen Palm fluxes (dynamics.F, ini_fields.F) |
271 |
|
o cleaned TAF keys iikey, idkey (dynamics.F, thermodynamics.F) |
272 |
|
o incorporated hooks for sponge layer code (external_forcing.F) |
273 |
|
|
274 |
|
checkpoint45d_post |
275 |
|
o import 2 fixes from release1_p5 (CG2D.h & packages_readparms.F) |
276 |
|
o use recip_dx*,recip_dy* instead of /dx*,/dy* in orlanski_E,W,N,S |
277 |
|
(affects plume_on_slope "cg2d" output) |
278 |
|
o add new parameter deltaTfreesurf for free-surface time stepping |
279 |
|
o exchange gT,gS when using staggerTimeStep & GM & Oly<4 |
280 |
|
o NONLIN_FRSURF: rescale gNm1 to get a better conservation with A-B |
281 |
|
o change phiMin in exp1 (to agree with documentation) |
282 |
|
o add new flags: |
283 |
|
- T,S forcing inside/outside Adams-Bashforth |
284 |
|
- turn off temp,salt advection and/or forcing |
285 |
|
- for each tracer, define internal flag for multiDimAdvection & A-B |
286 |
|
o monitor: change definition of KE to check conservation of Energy |
287 |
|
|
288 |
|
checkpoint45c_post |
289 |
|
o Removed f90 comments ("!") |
290 |
|
o Modified initialisations to enable exf package for MITgcm |
291 |
|
(without invoking ECCO_PACKAGE). |
292 |
|
o modifications to gradient check package (Martin Losch) |
293 |
|
- enable centered differences |
294 |
|
- modified format of standard output |
295 |
|
|
296 |
|
checkpoint45b_post |
297 |
|
o cleaned exchange calls in cg2d |
298 |
|
o Included CPP option ALLOW_AUTODIFF_MONITOR in forward_step |
299 |
|
o included CPP option SINGLE_LAYER_MODE in thermodynamics |
300 |
|
to configure barotropic setup (Martin Losch) |
301 |
|
o moved some initialisations within bi,bj loop in dynamics |
302 |
|
(TAF flow dependency) |
303 |
|
o in initialise_varia, moved call packages_init_variables |
304 |
|
before convective_adjustment_init |
305 |
|
(corresponding adjustment of TAF tape initialisation). |
306 |
|
o in calc_diffusivity/calc_viscosity extended domain to full overlap. |
307 |
|
|
308 |
|
checkpoint45a_post |
309 |
|
o compute Rho over the whole domain (extend iMin,jMin indices) |
310 |
|
o specific exchange for X-slice domain (case Ny=1) |
311 |
|
o argument futureIter of S/R obcs_calc missing in exp4 & internal_wave /code |
312 |
|
o solve_for_pressure : set cg2d_x to zero where OBCS are applied |
313 |
|
|
314 |
|
checkpoint44h_post = checkpoint45 |
315 |
|
o Fix to restore differentiability: |
316 |
|
place at which packages_init_variables is called in |
317 |
|
initialise_varia is crucial for reverse control flow. |
318 |
|
o modifications in GMRedi to enable stable adjoint |
319 |
|
(corresponding modif's in thermodynamics, dynamics) |
320 |
|
o added missing hook for sponge layer code in external_forcing |
321 |
|
o modified test cost functions |
322 |
|
o storing in gad_advection (plus removal of write(0,*)! |
323 |
|
o missing headers TR1.h added in convective_adjustment routines |
324 |
|
o errorMessageUnit set non-equal zero in eeboot |
325 |
|
(conflict on some platforms, e.g. IBM SP3) |
326 |
|
o modified carbon verif. |
327 |
|
|
328 |
|
checkpoint44h_pre |
329 |
|
o vertical grid option: allow to put the Interface at the middle between |
330 |
|
2 cell-centers ; replace delZ in pkg/kpp by drF. |
331 |
|
o GM Advective form: Tracers are advected using the residual transport (= |
332 |
|
Euler+GM-bolus); set param GM_AdvSeparate=T to return to previous form. |
333 |
|
|
334 |
|
checkpoint44g_post |
335 |
|
o fix surface correction term in multiDimAdvection (affects plume_on_slope |
336 |
|
results) ; ifdef missing in thermodynamics.F |
337 |
|
|
338 |
|
checkpoint44f_pre,checkpoint44f_post |
339 |
|
o added PTRACERS package |
340 |
|
This allows an arbitrary number of passive tracers to be integrated |
341 |
|
forward simultaneously with the dynamicaly model. |
342 |
|
+ Implemented so far: |
343 |
|
- basic forward algorithm (time-stepping, advection, diffusion, convection) |
344 |
|
- I/O and checkpointing |
345 |
|
- GM/Redi *but* using the GM/Redi coefficient of Salt |
346 |
|
+ Not implemented so far: |
347 |
|
- KPP |
348 |
|
- OBCS |
349 |
|
+ No specific example supplied (yet) but global_ocean.90x40x15 has the |
350 |
|
necessary data.ptracer file. Simply use -enable=ptracers and uncomment |
351 |
|
line in data.pkg. |
352 |
|
+ This package is disabled by default since it increases storage. |
353 |
|
|
354 |
|
|
355 |
|
o testscript: added -cleanup option |
356 |
|
o Shap_filt: a) new shap_filt S/R to use no-slip BC with S2 filter |
357 |
|
b) enable to filter 2D fields. |
358 |
|
|
359 |
|
checkpoint44e_pre,post |
360 |
|
o re-wrote convective adjustment to use pre-calculated weights |
361 |
|
Affects: model/src/convective_adjustment.F, convective_adjustment_ini.F |
362 |
|
Adds: model/src/convective_weights.F, convectively_mixtracer.F |
363 |
|
Original calls to convect() still in place. |
364 |
|
|
365 |
|
chkpt44d_pre,post |
366 |
|
o added missing EXCLUDE_MONITOR flags |
367 |
|
o changed "e" to "_d" in gmredi_slope_limit, gmredi_slope_psi |
368 |
|
(incompatible typ in MIN/MAX expressions caused problems |
369 |
|
on IBM SP3) |
370 |
|
o in genmake added variable MAKEDEPEND |
371 |
|
plus resetting for case SunOS |
372 |
|
o added timer_stats.c routine for IBM SP3 |
373 |
|
o removed variables in dynamics |
374 |
|
!!! I forgot to create tag chkpt44d_pre !!! |
375 |
|
!!! But chkpt44d_pre = chkpt44c_post !!! |
376 |
|
|
377 |
|
o real fresh water flux implemented with non-linear free-surface. |
378 |
|
|
379 |
|
chkpt44c_pre,post |
380 |
|
o few fix (mask in shap_s2, EmPmR in external_field_load, |
381 |
|
USE_NATURAL_BCS in solve_for_P); |
382 |
|
o add arguments myIter & myTime to S/R obcs_calc & solve_for_P |
383 |
|
|
384 |
|
checkpoint44b_pre/_post |
385 |
|
o merge of relevant stuff from the ecco-branch: |
386 |
|
- genmake: removed $S64 overwrite for case SunOS |
387 |
|
- pkg/exf: update and corrections for field swapping and obcs |
388 |
|
- pkg/ecco: parameter lists for the_model_main, the_main_loop |
389 |
|
harmonized between ECCO and MITgcm |
390 |
|
- pkg/autodiff: added flow directives for obcs, mdsio_gl_slice |
391 |
|
updated checkpointing_lev... lists for obcs |
392 |
|
- model/src: minor changes in forward_step, plot_field |
393 |
|
added directive for divided adjoint in the_main_loop |
394 |
|
- pkg/mdsio: added mdsio_gl_slice |
395 |
|
|
396 |
|
o check parameters & config (chkpt44a_pre,post) |
397 |
|
o OBC and NonLin_FrSurf. |
398 |
|
|
399 |
|
checkpoint44 |
400 |
|
o fix bug in mom_vi_del2uv |
401 |
|
o select when filters are applied ; add options to zonal_filter (data.zonfilt) |
402 |
|
o gmredi: fix Pb in the adiabatic form ; add options (.e.g. Bolus advection) |
403 |
|
o update AIM experiments (NCEP input files) |
404 |
|
o improve and extend diagnostics (Monitor, TimeAve with NonLin-FrSurf) |
405 |
|
o added some stuff for AD |
406 |
|
|
407 |
|
checkpoint43 |
408 |
|
o added two new advection tests |
409 |
|
o added global 4x4 degree ocean test |
410 |
|
o added test of H&S94 with free surface on cube |
411 |
|
o some final mods for gradient check, adjoint,... |
412 |
|
|
413 |
|
checkpoint42 |
414 |
|
Fixed AD-related problems: |
415 |
|
o Store directives up-to-date with re-arranged Adams-Bashforth |
416 |
|
(mainly thermodynamics.F) |
417 |
|
o New store directives for multi-dim. advection schemes |
418 |
|
* new CPP flag ALLOW_MULTI_DIM_ADVECTION |
419 |
|
* new common block and key passkey |
420 |
|
(mainly gad_advection.F) |
421 |
|
o Modified store directives for split of dynamics/thermodynamics |
422 |
|
for the case ALLOW_KPP |
423 |
|
o Cleaned argument list for timestep_tracer.F |
424 |
|
|
425 |
|
checkpoint41 |
426 |
|
o re-formatted/added comments in prologues of key subroutines |
427 |
|
to work with protex |
428 |
|
|
429 |
|
checkpoint40 |
430 |
|
o moved calc_gs, calc_gt and calc_tr1 back to model/src |
431 |
|
o added FLT (floats) package from Arne along with demo. experiment |
432 |
|
o re-arranged Adams-Bashforth scheme for tracers to allow easier |
433 |
|
implementation of forward-in-time forcing terms |
434 |
|
o more mods for non-linear free-surface including new variable |
435 |
|
o modified multi-dim method to work on cube |
436 |
|
o bug fix in KPP (from DM) |
437 |
|
|
438 |
|
checkpoint40pre9 |
439 |
|
o added DST advection schemes |
440 |
|
o fix MPI bug on SGI's |
441 |
|
o modifications for non-linear free-surface |
442 |
|
o fixed conditionals for fixed flow/passive tracer only mode |
443 |
|
o added tension/strain form of stress tensor |
444 |
|
o changes to commenting of arguments for TAF/TAMC |
445 |
|
o added/updated many matlab scripts |
446 |
|
o added multi-dimensional advection method for tracers |
447 |
|
|
448 |
|
checkpoint40pre8 |
449 |
|
o new package rw/ |
450 |
|
Added method for dumping intermediate local arrays: |
451 |
|
mdsio_writetile - same as mdsio_writefield except works from inside |
452 |
|
bi,bj loop |
453 |
|
mdsio_writelocal - same as mdsio_writetile except works for local |
454 |
|
arrays |
455 |
|
write_local_r? - higher-level wrapper for mdsio_writelocal |
456 |
|
Controlled by diagFreq. Defaults to zero (ie. no dumps) |
457 |
|
Example given at end of mom_vecinv.F that dumps some local arrays. |
458 |
|
o replaced calc_mom_rhs.F by mom_fluxform.F/mom_vecinc.F to |
459 |
|
control flux-form versus vector invariant code. |
460 |
|
Switch is runtime flag vectorInvariantMomentum (defaults to false) |
461 |
|
or genmake -DISABLE ... |
462 |
|
o TAMC-related update to cope with mom_fluxform.F/mom_vecinc.F |
463 |
|
(additional storing is now needed for case vecinv; |
464 |
|
currently nothing implemented to switch this storing off |
465 |
|
in case fluxform is used since no CPP flag available). |
466 |
|
|
467 |
|
checkpoint40pre7 |
468 |
|
o bug fix in dynamics.F |
469 |
|
calc_viscosity called before k=1,Nr loop. |
470 |
|
o added some #ifdef's ALLOW_PASSIVE_TRACER |
471 |
|
o cleaned initialisations in ctrl_map_ini |
472 |
|
related to TAF |
473 |
|
|
474 |
|
checkpoint40pre6 |
475 |
|
o split dynamics into thermodynamics & dynamics |
476 |
|
o corresponding split of calc_diffusivity into |
477 |
|
calc_diffusivity, calc_viscosity |
478 |
|
(kpp_calc_diff, kpp_calc_visc). |
479 |
|
o Added diffkr, kapgm to control variables |
480 |
|
o bug fix in gmredi_slope_limit for ldd97 |
481 |
|
|
482 |
|
checkpoint40pre5 |
483 |
|
o proper initialization of Adams-Bashforth time stepping |
484 |
|
+ uses forward step for very first step of model (iter=0) |
485 |
|
o re-generated *all* output since almost all output is modified by this |
486 |
|
change (sorry but we've been putting this one off for about 2 years) |
487 |
|
+ natl_box, aim.5l_Equatorial_Channel and aim.5l_zon-ave were |
488 |
|
*not* regenerated since they are already out of date and I don't |
489 |
|
want to create the impression that they are suddenly working. |
490 |
|
o removed KAP stuff for DEC's |
491 |
|
o modified/added many .cvsignore files in */code and */input |
492 |
|
o added new expt "solid-body.cs-32x32x1" |
493 |
|
+ this uses an alternative form of ini_curvilinear_grid.F which |
494 |
|
is not necessarily any better than the current one... |
495 |
|
|
496 |
|
checkpoint40pre4 |
497 |
|
o added re-initialisations of utrans,vtrans,rtrans in |
498 |
|
calc_common_factors which are crucial for TAF. |
499 |
|
o extended iMin,jMin range for calc_common_factors, |
500 |
|
calc_diffusivity |
501 |
|
o added option ALLOW_PASSIVE_TRACER |
502 |
|
o a few minor updates and fixes |
503 |
|
|
504 |
|
checkpoint40pre3 |
505 |
|
o created calc_gtr1.F for gad package |
506 |
|
o split the_main_loop and re-introduced forward_step |
507 |
|
o some recomputations of nIter0 for TAMC |
508 |
|
|
509 |
|
checkpoint40pre2 |
510 |
|
o Updated to latest branch of AJA, JMC |
511 |
|
o Added gradient check package |
512 |
|
o Added routines for passive tracer handling |
513 |
|
o Added carbon sequestration sensitivity experiment |
514 |
|
|
515 |
|
checkpoint39 |
516 |
|
o Modifications and fixes to support TAMC |
517 |
|
(mainly missing or wrong directives) |
518 |
|
o Corrected common blocks when fooling TAMC to produce adjoint I/O |
519 |
|
in addummy_in_stepping |
520 |
|
o Updated verification/vero/ (bug fixes in code/, correct pickups in input/) |
521 |
|
o Enabled "sliced" (xz-,yz-) exchanges, needed for obcs. |
522 |
|
o Included following packages: |
523 |
|
cal: calendar |
524 |
|
exf: external forcing |
525 |
|
ecco: essential routines to control use of cal, exf |
526 |
|
(NB: this is not the full ECCO environment) |
527 |
|
o disabled some packages by default in genmake |
528 |
|
|
529 |
|
checkpoint38 |
530 |
|
o see notes_c37_adj.txt |
531 |
|
o corrected wrong layer index in external_forcing_t |
532 |
|
o replaced the_main_loop/forward_step by the_main_loop |
533 |
|
o removed forceoutput from write_state |
534 |
|
(replaced criteria for initial/final state) |
535 |
|
o added runtime flag for pre-C35 pickups: usePickupBeforeC35 |
536 |
|
|
537 |
|
checkpoint37 |
538 |
|
o mdsio.F routine split and packaged in pkg/mdsio |
539 |
|
(mdsio.F and mdsio_gl.F still exist in eesupp/src but are not |
540 |
|
used unless genmake -disable=mdsio) |
541 |
|
o allow a more precise surface-geopotential gradient for the atmosphere |
542 |
|
taking into account the change in space of Alpha (=1/rho) at the surface. |
543 |
|
o variable "etaN" (replace cg2d_x) is the surface R-anomaly [r unit] |
544 |
|
(ocean: sea surface height [m] / atmos: surface pressure anomaly [Pa]) |
545 |
|
o Change units: Phi_Hydrostatic (=phiHyd), Phi_Surface(=cg2d_x) and |
546 |
|
Phi_NonHydrostatic (=cg3d_x), all have now the dimension of a potential |
547 |
|
[m2/s2] (pressure/rhoConst in the ocean ; Geo-Potential in the atmosphere). |
548 |
|
o Time-Average diagnostics defined as a pseudo package "timeave". |
549 |
|
o pickup files : old pickup (before c35) did not work with new code (c35 |
550 |
|
and after) unless activating 2 commented lines in initialise-varia. |
551 |
|
o Non-Hydrostatic pickup file Pb fixed. |
552 |
|
o bug with Rigid-lid version (since c35) fixed. |
553 |
|
|
554 |
|
checkpoint36 |
555 |
|
o implement Crank-Nickelson barotropic time-stepping |
556 |
|
o recover Time-Average diagnostics |
557 |
|
o fix Pb in checkpoint (PR#19) |
558 |
|
o AIM become a standard package ; clean unused CPP_kees |
559 |
|
|
560 |
|
checkpoint35 |
561 |
|
o subtantial rewrite of dynamics.f |
562 |
|
o allows staggered time-stepping |
563 |
|
o packaged obcs, aim (Atmospheric Physics), zonal_filt, shap_filt |
564 |
|
o added bottom drag (linear + quadratic) |
565 |
|
|
566 |
|
checkpoint34 |
567 |
|
o Modified exchange routines for correct reverse simulation. |
568 |
|
o Added verification output to comply with bug fix in impldiff.F |
569 |
|
as of c33 (exp2, natl_box). |
570 |
|
o Corrected store directives in dynamics.F |
571 |
|
o Corrected array boundaries in impldiff.F |
572 |
|
o Corrected array dimensions in gmredi. |
573 |
|
Added temporary scalars to avoid storage by TAMC (gmredi, kpp). |
574 |
|
o Added routine GLOBAL_SUM_INT to global_sum.F |
575 |
|
o Added mdsio_gl.F for ECCO purposes. |
576 |
|
|
577 |
|
checkpoint33 |
578 |
|
o Fixed bug in lower BC in impldiff.F |
579 |
|
o Fixed confusion about units of forcing arrays in FFIELDS.h |
580 |
|
namely Fu,Fv,Qnet,Qsw,EmPmR: |
581 |
|
- Removed verification/natl_box/code/external_fields_scale.F |
582 |
|
(did not differ from that in model/src) |
583 |
|
- Changed units of fu,fv,Qnet,Qsw,EmPmR back to proper units |
584 |
|
(see FFIELDS.h for description) |
585 |
|
- Scale fu,fv,Qnet,Qsw,EmPmR when used in external_forcing_surf.F, |
586 |
|
kpp_calc.F and kpp_transport_t.F |
587 |
|
- Removed model/src/external_fields_scale.F and calls to it |
588 |
|
- verification/natl_box uses flux data with "atmospheric" sign so |
589 |
|
a special version of external_fields_load.F is used to |
590 |
|
change the data as it's read in. This way, the arrays |
591 |
|
have the right units and signs at all times tha a user could |
592 |
|
possibly use them. |
593 |
|
o Corrected genmake to delete a temporary file during configuration |
594 |
|
|
595 |
|
checkpoint32 |
596 |
|
o Scaling of forcing fields moved from external_forcing.F to |
597 |
|
external_fields_scale.F, called directly after loading fields. |
598 |
|
o Surface relaxation terms added to surface tendency fields in |
599 |
|
extra subroutine. |
600 |
|
o impldiff.F changed to avoid storing by TAMC of huge intermediate |
601 |
|
fields. |
602 |
|
o Changes in GMRedi to comply with packaging structure. |
603 |
|
o Modifications and bug fixes in KPP package. |
604 |
|
o Verification experiment for KPP added (verification/nat_box/) |
605 |
|
|
606 |
|
checkpoint31 |
607 |
|
o Packages interface brought in line with defined standard |
608 |
|
-> new routines packages_...F |
609 |
|
-> changes in initialise_fixed.F, initialise_varia.F |
610 |
|
-> enables initialisation interface for ECCO package |
611 |
|
o Rescaling of external forcing fields immediately after read |
612 |
|
(removed from external_forcing.F) |
613 |
|
-> new: surface_fields_scale.F |
614 |
|
-> affects ini_foorcing.F, external_fields_load.F |
615 |
|
o Computation of surface tendencies before tendency updates. |
616 |
|
-> new: external_forcing_surf.F |
617 |
|
o Change of variable names useGMRedi, useKPP, useECCO |
618 |
|
o Further changes in dynamics.F, impldiff.F for TAMC |
619 |
|
o Tested for experiments 0, 2, 4. |
620 |
|
- not yet tested for GMRedi. |
621 |
|
- not yet tested for KPP. |
622 |
|
|
623 |
|
checkpoint30 |
624 |
|
o Updates for OBCs |
625 |
|
o New experiment for test OBCs (internal_wave) |
626 |
|
o New output.txt in verification/exp4/results/ |
627 |
|
o Updates for KPP: calculation of viscosity corrected, new variables |
628 |
|
kpp_dumpFreq and kpp_taveFreq |
629 |
|
o CPP'd calls to system routines for TARGET_CRAY_VECTOR (P.H.) |
630 |
|
and entry for T90 in genmake |
631 |
|
|
632 |
|
checkpoint29 |
633 |
|
o packaged KPP and GM/Redi |
634 |
|
|
635 |
|
checkpoint28 |
636 |
|
o Corrected initialization of hFacMin for difference vertical coordinates |
637 |
|
o Modified calculation of partial cell thickness to more robust algorithm |
638 |
|
o config_summary.F: replaced write_1D_... by write_0D_... |
639 |
|
read_write.F: added subroutines write_0D_... |
640 |
|
o routines calc_common_factors, calc_gs, calc_gt, calc_phi_hyd, find_rho: |
641 |
|
included some initialisations required for TAMC. |
642 |
|
o routine calc_isoslopes: modified sqrt computation to make it |
643 |
|
differentiable for zero argument. |
644 |
|
o routines dynamics, impldiff: |
645 |
|
included store directives and tape key computations required for TAMC. |
646 |
|
o ALL TAMC related changes are between ifdef ALLOW_AUTODIFF_TAMC |
647 |
|
|
648 |
|
checkpoint27 |
649 |
|
o fixed calc_isoslopes() to so that cancellation of terms in tensor |
650 |
|
formulation works properly with variable resolution. |
651 |
|
o restructured the calling tree between the_model_main() and dynamics() |
652 |
|
o split initialise() into initialise_fixed() and initialise_varia() |
653 |
|
o introduced initialization file pSurfInitFile for free surface (cg2d_x) |
654 |
|
in new routine ini_psurf() |
655 |
|
|
656 |
|
checkpoint26 |
657 |
|
o moved some #ifdef in optional routines to encompass everything |
658 |
|
except the argument declarations. This was using up some memory. |
659 |
|
o fixed macros like _ADT to work with both versions of Linux cpp |
660 |
|
o cleaned up some unused and uninitialized variables |
661 |
|
(helps when debugging with strict compile options) |
662 |
|
o split up CPP_EEOPTIONS.h into options and macros |
663 |
|
(macros are now in eesupp/inc/CPP_EEMACROS.h) |
664 |
|
o patch for 2 processor JAM mode |
665 |
|
|
666 |
|
checkpoint25 |
667 |
|
o updates for OBCs and NH |
668 |
|
o dramatically reduced memory usage |
669 |
|
o added "JAM" routines for use on Artic network |
670 |
|
o parameterization of convection by implicit vertical diffusion |
671 |
|
o added a diagnostic of convective index |
672 |
|
o moved call to calc_divg_hat() from dynamics() to solve_for_pressure() |
673 |
|
(this is was partly for OBCs and NH but also in anticipation of |
674 |
|
an explicit free-surface option) |
675 |
|
o initial state input files for U and V |
676 |
|
|
677 |
|
checkpoint24 |
678 |
|
o Removed some spurious "REAL"s |
679 |
|
o Updated KPP to latest version from JPL |
680 |
|
o Added a COS(latitude) in horizontal viscosity |
681 |
|
|
682 |
|
checkpoint23 |
683 |
|
o Added del^4 diffusion for salt and temperature |
684 |
|
o Multiple minor fixes (implicit none's, arguments to global_sum, etc.) |
685 |
|
|
686 |
|
checkpoint22 |
687 |
|
o Made many small changes: implicit none, format(i5), etc... |
688 |
|
o Introduced KPP mixing scheme (compiles and runs but *not* tested) |
689 |
|
|
690 |
checkpoint21 |
checkpoint21 |
691 |
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) |
692 |
o Split-up mid-level I/O routines |
o Split-up mid-level I/O routines |