4 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
5 |
============================== |
============================== |
6 |
|
|
7 |
|
o pkg/obcs: |
8 |
|
- change default setting of OBCS_UVICE_OLD from defined to undef; |
9 |
|
- use new parameter OB_indexNone for null index value (instead of hard-coded 0) |
10 |
|
in all pkg/obcs code and in other pkgs; remove k loop in obcs_check_depths.F; |
11 |
|
- pass OB_indexNone as (new) argument to S/R CTRL_MASK_SET_[X,Y]Z; |
12 |
|
- change OB null index value (OB_indexNone) to -99 ; this fix the problem |
13 |
|
with Southern & Western OB at a tile edge (at least in a simple set-up). |
14 |
|
o pkg/obcs: |
15 |
|
- change settings of OB index in case OB coincides with tile edge (shared by |
16 |
|
2 tiles), so that common OB location is registered by both tiles (instead of |
17 |
|
only 1). However, for Southern and Western OB, index value = 0 is not |
18 |
|
recognized as valid OB index (=default null value) and needs to be fixed. |
19 |
|
- modify accordingly OBCS_BALANCE_FLOW and OBCS_MONITOR (avoid double counting) |
20 |
|
- new parameter "OB_indexNone" for null index value (for now, still zero) |
21 |
|
to replace hard-coded null index value of zero. |
22 |
|
o pkg/exch2: |
23 |
|
- improve W2_USE_E2_SAFEMODE (faster) and W2_E2_DEBUG_ON: |
24 |
|
always check buffer-size (but outside the loop); move checking for valid |
25 |
|
index from W2_USE_E2_SAFEMODE to W2_E2_DEBUG_ON; in W2_E2_DEBUG_ON, |
26 |
|
print each tile and point connection only if |W2_printMsg|>= 2 and 3. |
27 |
|
o pkg/ctrl (pack & unpack): |
28 |
|
- in ctrl_pack.F & ctrl_unpack.F, use the same local file-name variable |
29 |
|
(locally defined) for different variables (+ remove the variable-specific |
30 |
|
file name in CTRL_GENARR.h). For now, only implemented for GENARR2/3D_CONTROL |
31 |
|
(to be extended to other local file names in ctrl.h). |
32 |
|
o pkg/flt: |
33 |
|
- add a FLT_Iter0 option (similar to PTRACERS_Iter0). |
34 |
|
o pkg/mdsio: |
35 |
|
- fix missing STOP in mdsio_readvec_loc.F (case ioUnit=0) when file is missing |
36 |
|
o tools/genmake2: |
37 |
|
- add similar dependences for target EXE_AD & EXE_FTL as we have for EXECUTABLE |
38 |
|
(added by Alistair in version 1.51 for building without make depend); |
39 |
|
- remove Header files list (H_SRC_FILES) from the dependence of targets |
40 |
|
ad_input_code.f & ftl_input_code.f (not clear why it was added in the first |
41 |
|
place in version 1.23): this save going through staf when unrelated header |
42 |
|
file has been modified. |
43 |
|
Note: for an intermediate target (e.g., ad_input_code.f or adtaf) without |
44 |
|
"make depend" step, still need to do "make links" before "make adtaf". |
45 |
|
o eesupp/src (+ few pkg src files) & genmake2: |
46 |
|
- finish to remove ALWAYS_USE_MPI in source code that TAF does not see; |
47 |
|
- no longer set ALWAYS_USE_MPI in Makefile (was done by genmake2). |
48 |
|
- improve genmake2 checking for ALLOW_${PKG} in CPP_OPTIONS.h files (distinguish |
49 |
|
upper/lower case ; account for space between "#" and undef/define) |
50 |
|
o pkg/flt (with exch2): |
51 |
|
- cleaning: fix so that it compiles; move remaining MPI calls out of flt_exch2.F; |
52 |
|
- simplify threads synchronisation; fix (many issues) multi-threaded with MPI. |
53 |
|
|
54 |
|
checkpoint63s (2012/09/05) |
55 |
|
o pkg/ecco: |
56 |
|
- remove #ifdef ALLOW_SMOOTH_CORREL3D brackets. |
57 |
|
- add more relevant #ifdef ALLOW_SMOOTH ones. |
58 |
|
- sort out useAtmWind, useSMOOTH, ctrlSmoothCorrel2D. |
59 |
|
o pkg/ctrl: |
60 |
|
- CTRL_OPTIONS.h : document CPPs related to smoothing; ALLOW_SMOOTH_CORREL2D/3D, |
61 |
|
ALLOW_CTRL_SMOOTH, ALLOW_SMOOTH_CTRL2D, ALLOW_SMOOTH_CTRL3D. |
62 |
|
- ctrl.h : document run time params doInitXX, doMainPack, doMainUnpack, doPackDiag, |
63 |
|
doSinglePrecTapelev (audodiff param!), ctrlSmoothCorrel2D (new) and |
64 |
|
ctrlSmoothCorrel3D (new). |
65 |
|
The added ctrlSmoothCorrel2D/3D replace the old ALLOW_SMOOTH_CORREL2D/3D CPP. |
66 |
|
- ctrl_readparms.F : initialise ctrlSmoothCorrel2D/3D and read from namelist. |
67 |
|
- ctrl_init.F ctrl_pack.F ctrl_unpack.F : switch off ustress/vstress |
68 |
|
controls when useAtmWind, and vice versa. |
69 |
|
- ctrl_map_ini_ecco.F : UVEL0, VVEL0, edtaux, edtauy cannot be smoothed as tracers. |
70 |
|
- ctrl_get_gen.F : replace #ifdef ALLOW_SMOOTH_CORREL2D with |
71 |
|
#ifdef ALLOW_SMOOTH, IF ctrlSmoothCorrel2D, IF useSMOOTH. |
72 |
|
- ctrl_map_ini_gen.F : replace #ifdef ALLOW_SMOOTH_CORREL3D with |
73 |
|
#ifdef ALLOW_SMOOTH, IF ctrlSmoothCorrel3D, IF useSMOOTH. |
74 |
|
- ctrl_set_pack_xy[z].F, ctrl_set_unpack_xy[z].F: |
75 |
|
replace ALLOW_SMOOTH_CORREL2D/3D brackets with ctrlSmoothCorrel2D/3D. |
76 |
|
o pkg/smooth: |
77 |
|
- remove un-necessary includes. |
78 |
|
- remove ALLOW_SMOOTH*D and ALLOW_SMOOTH_CORREL*D brakets. |
79 |
|
Those CPP options were never defined, and not necessary. |
80 |
|
o model/src: |
81 |
|
- fix CPP brackets in ini_mixing.F (GM_background_K belongs to gmredi). |
82 |
|
- standard treatment of useSMOOTH. |
83 |
|
o pkg/autodiff: |
84 |
|
- add LOGICAL useGGL90inAdMode, useGGL90inFwdMode, useSALT_PLUMEinAdMode, |
85 |
|
useSALT_PLUMEInFwdMode allowing to switch off ggl90 and salt_plume adjoint. |
86 |
|
o pkg/exch2: |
87 |
|
- add "if usingMPI" where it was missing (in exch2_rx?_cube*.template) |
88 |
|
+ start to remove ALWAYS_USE_MPI in source code that TAF does not see. |
89 |
|
- make new S/R W2_MAP_PROCS out of w2_eeboot.F to setup exch2 CommonFlag |
90 |
|
and tiles to process mapping. |
91 |
|
- re-arrange 3 header files (W2_EXCH2_PARAMS.h, _BUFFER.h & _TOPOLOGY.h) |
92 |
|
moving /EXCH2_FILLVAL_RX/ & /W2_EXCH2_COMMFLAG/ to W2_EXCH2_TOPOLOGY.h; |
93 |
|
rename "exch2_tProc" to "W2_tileProc" and move seting of e2FillValue_RX |
94 |
|
from w2_eeboot.F to w2_readparms.F. |
95 |
|
o eesupp/src: |
96 |
|
- set exchNeedsMemsync & exchUsesBarrier according to "usingMPI" (instead of |
97 |
|
ALWAYS_USE_MPI) + moved from ini_communication_patterns.F to eeset_parms.F; |
98 |
|
also move setting of exchCollectStatistics from exch_init.F to eeset_parms.F |
99 |
|
- in eedie.F and all_proc_die.F, always call MPI_FINALIZE when ALLOW_USE_MPI |
100 |
|
is defined (remove condition on usingMPI) to match cases where MPI_INIT |
101 |
|
has been called. |
102 |
|
- fix bug I put in previous modif of exch_sm_3d_rx.template (not much used); |
103 |
|
- change default (non-MPI) value of myPid (from 1 to 0) to match MPI settings |
104 |
|
(so that on 1 proc, get same myPid independently of usingMPI) |
105 |
|
- add a stop if multi-procs but usingMPI=F ; stop also if usingMPI=T and |
106 |
|
useCubedSphereExchange with EXCH-1 (safer). |
107 |
|
- add "if usingMPI" where it was missing (+ start to remove ALWAYS_USE_MPI). |
108 |
|
o pkg/ggl90: |
109 |
|
- fix bug in filling corner mask array "mskCor" (FILL_CS_CORNER_TR_RL call) |
110 |
|
used in ALLOW_GGL90_SMOOTH code. |
111 |
|
o pkg/flt: |
112 |
|
- fix typo in FLT_WITHOUT_Y_PERIODICITY (the effect was to ignore stopping |
113 |
|
floats at the Southern end of the domain). |
114 |
|
o eesupp/inc: |
115 |
|
- add a list of CPP-options which can be set by Makefile (DEF_IN_MAKEFILE.h); |
116 |
|
but not used (and not to be used) anywhere, just for documentation. |
117 |
|
o pkg/autodiff, exf, ptracers & seaice (AD & TL output): |
118 |
|
- comment out the very nasty "writeBinaryPrec = writeStatePrec" line, since: |
119 |
|
a) can damage previously written records in any other multi-record files. |
120 |
|
b) if needed, can easily do it cleanly, by calling S/R WRITE_REC_3D_RL/S. |
121 |
|
o tools/genmake2: |
122 |
|
- define variable MAKE within Makefile and use it in place of 'make'; |
123 |
|
- update AD_CONFIG.h (if needed) before building any of the executable |
124 |
|
(forward, adjoint or tangent-lin); this involves a 2nd invocation of make; |
125 |
|
and add the usual #ifndef/#define AD_CONFIG_H in AD_CONFIG.h; |
126 |
|
- always print usual warning comment 'this file is automatically generated' |
127 |
|
from script convert_cpp_cmd2defines (+ remove this warning from genmake2). |
128 |
|
o pkg/cheapaml: |
129 |
|
- updated to correspond to the code used in accepted paper (Nico Wienders). |
130 |
|
o pkg/exf: |
131 |
|
added run time switch useAtmWind to replace ALLOW_ATM_WIND cpp switch. |
132 |
|
ALLOW_ATM_WIND now just sets the useAtmWind default (see exf_readparms.F) |
133 |
|
and force defines ALLOW_BULKFORMULAE (EXF_OPTIONS.h). |
134 |
|
o pkg/exf, autodiff, ctrl, ecco and seaice: |
135 |
|
remove ALLOW_ATM_WIND brackets, or replace them with useAtmWind ones. |
136 |
|
o pkg/ctrl, ecco: |
137 |
|
allow to compile both ALLOW_U/VSTRESS_CONTROL and ALLOW_U/VWIND_CONTROL. |
138 |
|
Depending on useAtmWind, one is inactive, and the other is active (see |
139 |
|
exf_getffields.F/exf_getsurfacefluxes.F). |
140 |
|
o tools/genmake2: |
141 |
|
- improve error msg when failing in 2nd test of check_for_broken_Ff |
142 |
|
- fix option '-makefile' short-name (-mf according to genmake2 -help) |
143 |
|
- source local config file "genmake_local" just after processing the genmake2 |
144 |
|
command arguments (previously done before processing arg). |
145 |
|
o pkg/grdchk: |
146 |
|
- add a call to TURNOFF_MODEL_IO for Tangent-Linear run (in this case, call |
147 |
|
to this routine from cost_final has been dropped in g_cost_final) |
148 |
|
- improve printed information (more explicit msg, no longer using "print *," |
149 |
|
fix fcpertminus printed value). |
150 |
|
o model/src: |
151 |
|
- move call to COST_INIT_VARIA out of packages_init_variables.F directly |
152 |
|
into initialise_varia.F (just after calling PACKAGES_INIT_VARIABLES): |
153 |
|
this prevents getting cost-funct reset in adinitialise_varia recomputations. |
154 |
|
o pkg/dic: |
155 |
|
- move initialisation of "freefe" array from from dic_biotic_forcing.F to |
156 |
|
fe_chem.F: this prevents TAF from dropping this initialisation in TLM code. |
157 |
|
o pkg/grdchk: |
158 |
|
- grdchk_main.F : added CALL GLOBAL_SUM_INT( ierr , myThid ) |
159 |
|
- grdchk_getadxx.F etc. : omit I/O if ierr.EQ.0 |
160 |
|
o pkg/autodiff/ctrl/grdchk : |
161 |
|
- attempt to sort out what works and what does not, |
162 |
|
with regard to I/O, and to warn/stop accordingly. |
163 |
|
o pkg/grdchk: |
164 |
|
- attempt to fix grdchk in the case of useSingleCpuIO. |
165 |
|
- "if ( myProcId .EQ. grdchkwhichproc .AND. ierr .EQ. 0 )" |
166 |
|
statements were moved from grdchk_main.F and into |
167 |
|
grdchk_getadxx.F grdchk_getxx.F grdchk_setxx.F. |
168 |
|
- a global sum is applied to record numbers icvrec/layer |
169 |
|
out of grdchk_loc so that they be passed to I/O routines. |
170 |
|
o pkg/smooth: |
171 |
|
- fix recomputation that appeared with nlfs : use smooth_recip_hFacC, |
172 |
|
smooth_hFacW, smooth_hFacS (copied at init_fixed) in 3d case. |
173 |
|
o verification/testreport: |
174 |
|
- add option '-tlm' for Tangent-Linear test (-adm,-oad,-tlm are all exclusive); |
175 |
|
- can also use compressed (gzip) reference output file; |
176 |
|
- change format (in pkg/grdchk) of TLM output (+ add forward gradient) |
177 |
|
to be easier to collect by testreport. |
178 |
|
- update existing TLM output (half out-of-date by more than 3 yrs). |
179 |
|
- report NETCDF & LAPACK settings (from genmake2) in output dir. |
180 |
|
- report run-time in each individual summary (previously only for AD test); |
181 |
|
for TLM test, add also TAF version. |
182 |
|
- add daily TLM test on (old) aces cluster + weekly test on baudelaire. |
183 |
|
|
184 |
|
checkpoint63r (2012/08/15) |
185 |
|
o tools/genmake2: |
186 |
|
- fix Makefile for target ftlall (+ update files to clean after make ftlall) |
187 |
|
o pkg/ecco/ecco_cost_final.F : |
188 |
|
- fix fc print statements (to stdout and costfunction0000) so |
189 |
|
that the printed value is always after the global sum. |
190 |
|
o pkg/seaice/seaice_dynsolver.F : |
191 |
|
- fix ice strength assymptote at SEAICE_area_max<1 |
192 |
|
(no change for the standard case of SEAICE_area_max=1) |
193 |
|
o some cleaning: |
194 |
|
- pkg source files: ensure that all *.F files from each pkg included the |
195 |
|
corresponding ${PKG}_OPTIONS.h file (except stand-alone src file which |
196 |
|
only includes CPP_EEOPTIONS.h); fix S/R CTRL_SMOOTH ; update AD experiment |
197 |
|
code_ad/*.F src files; add new option-file in pkg/admtlm + corresponding |
198 |
|
includes; |
199 |
|
- model/src: add PACKAGES_CONFIG.h wherever ALLOW_AUTODIFF[_TAMC] is used. |
200 |
|
- ALLOW_DIFFKR_CONTROL: add a stop in S/R CTRL_CHECK in case ALLOW_3D_DIFFKR |
201 |
|
is undef; also define /DYNVARS_DIFFKR/ common block in DYNVARS.h when |
202 |
|
ALLOW_DIFFKR_CONTROL is defined, to allow to compile and reach this stop. |
203 |
|
- update option-file CPP_OPTIONS.h of all AD experiments: |
204 |
|
* no longer include PACKAGES_CONFIG.h ; |
205 |
|
* includes ECCO_CPPOPTIONS.h if PACKAGES_CONFIG.h has been included ; |
206 |
|
* def/undef ALLOW_3D_DIFFKR independently of ALLOW_DIFFKR_CONTROL (before |
207 |
|
including ECCO_CPPOPTIONS.h). |
208 |
|
o pkg/cost, pkg/ctrl, pkg/ecco option-file: |
209 |
|
- clean-up and split content of old pkg/ecco/ECCO_CPPOPTIONS.h file into |
210 |
|
separated pkg option-file (done according to old doc from ECCO_CPPOPTIONS.h, |
211 |
|
not necessary accurate). |
212 |
|
- add a template for option-file ECCO_CPPOPTIONS.h, mostly empty, in |
213 |
|
pkg/autodiff, i.e., the 1rst pkg to compile for adjoint built, with lines to |
214 |
|
deliberately break compilation (a remainder that it needs to be customized) |
215 |
|
- in pkg/ecco, includes ECCO_OPTIONS.h instead of COST_OPTIONS.h |
216 |
|
- rename COST_CPPOPTIONS.h to COST_OPTIONS.h |
217 |
|
- rename CTRL_CPPOPTIONS.h to CTRL_OPTIONS.h |
218 |
|
o tools/genamke2: |
219 |
|
- check for lapack. |
220 |
|
o pkg/cost and pkg/ecco : attempt to sort out some of the limiting intricacies. |
221 |
|
- pkg/cost.cost_final.F : allow for use of pkg/ecco and pkg/cost cost functions |
222 |
|
jointly. The exclusive ifdefs were removed, to add costs terms (when defined) |
223 |
|
from both packages. |
224 |
|
- the fc global sum in pkg/ecco/ecco_cost_final.F was removed accordingly. |
225 |
|
- for objf_atl we use either the pkg/ecco or the pkg/cost term depending on CPPs. |
226 |
|
- pkg/ecco and model/src : add CPP brackets around pkg/cost and pkg/ctrl includes |
227 |
|
of "CTRL_SIZE.h" "ctrl.h" "ctrl_dummy.h" "optim.h" / "cost.h" accordingly. |
228 |
|
o pkg/cost : add comments to explain special and/or untested setups. |
229 |
|
o pkg/exf: |
230 |
|
- use EXF_OPTIONS.h to set pkg/exf options except if done in ECCO_CPPOPTIONS.h |
231 |
|
(previously, ALLOW_AUTODIFF_TAMC was controling this switch). |
232 |
|
o verification/*/code_ad: |
233 |
|
mimimal cleaning of AD experiments ECCO_CPPOPTIONS.h option-file: |
234 |
|
- reccord when this file is included (define ECCO_CPPOPTIONS_H) |
235 |
|
- remove the most obvious obsolete features (e.g., #define ALLOW_COST not |
236 |
|
allowed since Oct 2003 in any option-file except in PACKAGES_CONFIG.h) |
237 |
|
o pkg/seaice : |
238 |
|
- added possibility to change ice strength/PRESS0 relationship to |
239 |
|
HEFF, according to the following run time parameters |
240 |
|
SEAICEpresPow0 :: HEFF exponent for ice strength below SEAICEpresH0 (integer, >0) |
241 |
|
SEAICEpresPow1 :: HEFF exponent for ice strength above SEAICEpresH0 (integer, >0) |
242 |
|
SEAICEpresH0 :: HEFF threshold for ice strength (m) |
243 |
|
o pkg/ggl90 : |
244 |
|
- added run time flag mxlSurfFlag to include the code that ensure |
245 |
|
mixing between first and second level (previously included as a comment) |
246 |
|
o verification/1D_ocean_ice_column/input_ad/data.autodiff : |
247 |
|
- add test of newly added dumpAdByRec=T option where |
248 |
|
we write all records to one file per variable. |
249 |
|
o verification/global_ocean.cs32x15/code_ad/CPP_OPTIONS.h : |
250 |
|
- switch to #define ALLOW_BALANCE_RELAX and ALLOW_BALANCE_FLUXES |
251 |
|
to check that this code is ok with taf and compiles in adjoint. |
252 |
|
o pkg/autodiff : |
253 |
|
- added run time param dumpAdByRec : ad dump format switch. |
254 |
|
If true then write all records to one file per variable. |
255 |
|
Else write one file per record (remains the default). |
256 |
|
- revise seaice_ad_dump.F and exf_adjoint_snapshots_ad.F accordingly. |
257 |
|
o pkg/seaice: |
258 |
|
- Added SEAICEuseTILT run time switch : if set to |
259 |
|
false then omit the tilt term in seaice dynamics. |
260 |
|
- Fix free drift calculation. |
261 |
|
o pkg/shelfice: |
262 |
|
When initializing SHELFICEDragQuadratic, set default to be |
263 |
|
consistent with shiCdrag in case SHELFICEuseGammaFrict is used, |
264 |
|
but retain option to set it independently |
265 |
|
(inconsistenty spotted by Veronique Dansereau) |
266 |
|
o eesupp/inc: |
267 |
|
- in EEPARAMS.h, add real-type constant for some frequently used simple number |
268 |
|
(0,1,2,1/2), with name explicitly referring to RS/RL type. |
269 |
|
o model/src: |
270 |
|
- move balancing of relaxation terms to new S/R balance_relax.F. |
271 |
|
o pkg/autodiff: |
272 |
|
re-write active I/O high-level (control) S/R: |
273 |
|
- rename S/R ACTIVE_READ/WRITE_RL to ACTIVE_READ/WRITE_3D_RL |
274 |
|
- add 1 more argument (useCurrentDir) to replace the "_LOC" 2nd set of S/R. |
275 |
|
- use standard pkg/mdsio I/O routines (MDS_READ/WRITE_SEC_XZ/YZ and |
276 |
|
MDS_READ/WRITE_FIELD); change loops to multi-threads. |
277 |
|
- update and merge active_file_control_slice.flow into active_file_control.flow |
278 |
|
- update the calls in lower level S/R (active_file.F, active_file_g.F and |
279 |
|
active_file_ad.F) keeping the argument list and function unchanged. |
280 |
|
- also update lower-level "_LOC" version (call same S/R with "useCurrentDir |
281 |
|
=.TRUE.") and remove high-level (control) _loc version (not called anymore). |
282 |
|
- new low-level active I/O routines, one for each type (RL/RS), for active var |
283 |
|
of various (XY,XZ,YZ) shape and level number (arguments: arrShape, myNr); |
284 |
|
fix ctrl_init_wet.F & ecco_cost_weights.F by calling the new S/R with the |
285 |
|
right type. |
286 |
|
o autodiff and RS=real*4: |
287 |
|
- call the right type (_RS instead of _RL) of S/R in few output S/R (pkg/exf |
288 |
|
ad-monitor and snapshots); disable arrType='RS' in mdsio_gl*.F ; |
289 |
|
- change argument type of 3-D mask argument to _RS to match type in the call |
290 |
|
(in seaice_cost*.F, ctrl_*_gen.F and *ctrl_bound.F); to be continued. |
291 |
|
o pkg/thsice: |
292 |
|
- re-anable balanceEmPmR when useSEAICE=T with useThSIce=T. |
293 |
|
- option to correct ocean fresh-water flux for imbalance of Atmos+Land |
294 |
|
fresh-water flux: uniform correction or proportional to local precip. |
295 |
|
o pkg/ctrl: |
296 |
|
- new header file CTRL_SIZE.h for parameters maxCtrlArr2D, maxCtrlArr3D |
297 |
|
- redefine CPP option ALLOW_BOTTOMDRAG_CONTROL to |
298 |
|
ALLOW_BOTTOMDRAG_CONTROL_NONGENERIC |
299 |
|
- verification/hs94.1x64x5/ tests ALLOW_GENARR{2D,3D}_CONTROL |
300 |
|
|
301 |
|
checkpoint63q (2012/07/30) |
302 |
|
o pkg/autodiff: |
303 |
|
note about monitor and snap-shot of ad-variables: |
304 |
|
a) need to apply ADEXCH before monitor or output of ad-vars to include |
305 |
|
overlap contributions (was done for adjDump but missing for monitor). |
306 |
|
b) since ADEXCH implies adding overlap values to interior array, machine |
307 |
|
truncation error is expected when adding an ADEXCH call in the middle of |
308 |
|
the time-stepping (although mathematically identical). |
309 |
|
modifs: |
310 |
|
- fix some incorrect type (RL/RS, UV/scalar) in ADEXCH calls and ad-vars |
311 |
|
declaration (adEddyPsiX,adEddyPsiY in adcommon.h). |
312 |
|
- add new run-time params (dumpAdVarExch & mon_AdVarExch) to select how to |
313 |
|
apply ADEXCH before monitor or snap-shot of ad-variables: |
314 |
|
=0 : no ADEXCH ; =1 : do ADEXCH ; =2 : apply ADEXCH on a local copy. |
315 |
|
- implement new method (copy+adexch) for snap-shot and monitor of main model |
316 |
|
ad-vars (initially not used by default -> output unchanged). |
317 |
|
- change default (set to 2) to use new method and update few output_adm.txt |
318 |
|
with accurate ad-monitor (global_ocean.90x40x15, tutorial_global_oce_biogeo |
319 |
|
and both tutorial_tracer_adjsens tests). |
320 |
|
- switch 2 global_ocean.cs32x15 AD-tests to AB-2 (seaice_dynmix & thsice) |
321 |
|
and turn off momDissip_In_AB in the 2 other (needed for stable AB-3). |
322 |
|
o pkg/ctrl: |
323 |
|
enable the use of a variable number (maxCtrlArr2D, maxCtrlArr3D) of |
324 |
|
generic control variables via |
325 |
|
#define ALLOW_GENARR2D_CONTROL and/or #define ALLOW_GENARR3D_CONTROL |
326 |
|
Only required customization should be S/R ctrl_map_ini_genarr.F |
327 |
|
(routine in repository contains example) |
328 |
|
o pkg/exf: |
329 |
|
- fix and refine checking of surface-relaxation setting when using EXF |
330 |
|
o verification (adjoint exp): |
331 |
|
- reduce length of simulation by half for experiment global_ocean.cs32x15 |
332 |
|
(std, seaice & seaice_dynmix) and exp tutorial_dic_adjoffline; update output. |
333 |
|
o pkg/mom_common (mom_u/v_bottomdrag), pkg/shelfice (shelfice_u/v_drag) |
334 |
|
- replace if statements KE.ne.0 with KE.gt.0 to avoid sqrt of negative numbers |
335 |
|
of KE that can occur with aggressive optimization |
336 |
|
+ put quadratic bottom drag calculation in separated loops (only done if |
337 |
|
using bottomDragQuadratic) |
338 |
|
o model/src: CG2D_NSA |
339 |
|
- clean-up: stop if cg2dNormaliseRHS=T and AUTODIFF |
340 |
|
and replace GLOBAL_SUM calls with GLOBAL_SUM_TILE. |
341 |
|
- in test-exp that uses CG2D_NSA (bottom_ctrl & tutorial_tracer_adjsens), |
342 |
|
change cg2d residual target to be expressed in Wunits (=> do not normalise |
343 |
|
RHS ; this fix global_max Pb with TAF); update output. |
344 |
|
o verification/tutorial_tracer_adjsens: |
345 |
|
improve parameter consistency (and get rid of warnings): |
346 |
|
- use deltaTFreeSurf=deltaTtracer (better with NonLin FreeSurf); |
347 |
|
- switch Temp Adv Scheme from 81 to 80 (no limiter since theta < 0 is physical) |
348 |
|
- use staggerTimeStep (for stability) in som81; use default rotation period. |
349 |
|
o pkg/exch2: |
350 |
|
- with empty facet: fix compact-format (W2_mapIO=1) definition (was previously |
351 |
|
reset to 0 = 1 long line in X) |
352 |
|
o pkg/mdsio: |
353 |
|
- fix reading of dimensions from meta file for large-size domain |
354 |
|
o model/src: |
355 |
|
- fix saltIsActiveTr setting for IdealGas eosType (eosType is 6 character long); |
356 |
|
- to refine CS-grid check, add logical flag "hasWetCSCorners": true if using |
357 |
|
Cubed-Sphere Exch with CS-corners inside the domain; |
358 |
|
make new S/R INI_GLOBAL_DOMAIN from code in ini_linear_phisurf.F to calculate |
359 |
|
globalArea and to set hasWetCSCorners. |
360 |
|
- add warning if usingCurvilinearGrid and momAdvection in flux-form (missing |
361 |
|
metric terms); stop if hasWetCSCorners=T and momAdvection in flux-form. |
362 |
|
- re-activate the stop if useCDscheme and hasWetCSCorners=T (previously if |
363 |
|
useCDscheme and useCubedSphereExchange). |
364 |
|
o pkg/seaice/seaice_growth.F : |
365 |
|
- compute the SEAICE_DISABLE_HEATCONSFIX term accordingly |
366 |
|
for the (.NOT.useRealFreshWaterFlux).OR.(nonlinFreeSurf.EQ.0) |
367 |
|
when the previous code was not correct. |
368 |
|
- in all cases diagnoze the boundary condition mismatch to SIaaflux |
369 |
|
- in the case of useRealFreshWaterFlux.AND.(nonlinFreeSurf.GT.0) add |
370 |
|
that term to QNET (essentially to set the melt/freeze water to 0degC). |
371 |
|
when SEAICEheatConsFix is true, so that the ocean+ice budget heat is closed. |
372 |
|
o verification/global_ocean.cs32x15/input.seaice : |
373 |
|
- add SEAICEheatConsFix in data.seaice, ensuring closed ocean+ice heat budget. |
374 |
|
- add diagnostics (commented out) for budget test using my matlab codes (gcmfaces). |
375 |
|
- update results (output.seaice.txt) accordingly. |
376 |
|
o verification/lab_sea/input/data.seaice : |
377 |
|
- remove SEAICEheatConsFix, which is now be tested |
378 |
|
in global_ocean.cs32x15. The presently coded fix |
379 |
|
is only consistent with real fresh water + nlfs. |
380 |
|
- update results (output.txt) accordingly. |
381 |
|
o pkg/salt_plume: added diagnotic for salt tendency (oceSPtnd). |
382 |
|
o AD_CONFIG.h file: |
383 |
|
- in genmake2: remove "#undef ALLOW_ECCO_OPTIMIZATION" from AD_CONFIG.h; |
384 |
|
- remove #include "AD_CONFIG.h" from all included OPTIONS files and |
385 |
|
add it explicitly in each fortran src file where it is needed. |
386 |
|
motivation: header file AD_CONFIG.h is generated/modified after "make depend" |
387 |
|
thus introducing potentially wrong dependency; therefore, better to include |
388 |
|
it only where it's really needed, and after any other included OPTIONS file |
389 |
|
(less tempting to bring conditional includes according to ALLOW_ADJOINT_RUN |
390 |
|
or ALLOW_TANGENTLINEAR_RUN). |
391 |
|
|
392 |
|
checkpoint63p (2012/07/06) |
393 |
|
o pkg/autodiff: |
394 |
|
- rename S/R INI_AUTODIFF to AUTODIFF_INIT_VARIA ; |
395 |
|
- create new header file "AUTODIFF_PARAMS.h" for pkg/autodiff parameters |
396 |
|
which are now read from new parameter file "data.autodiff" (inAdExact |
397 |
|
previously in "data", use{KPP/GMRedi/SEAICE}inAdMode previously in data.pkg) |
398 |
|
- add file "data.autodiff" in all verification/*/input_ad/ directories. |
399 |
|
- add run-time parameter to turn On/Off SEAICEuseFREEDRIFT in adjoint/forward |
400 |
|
mode (used in global_ocean.cs32x15.seaice_dynmix test exp.) |
401 |
|
o verification/tutorial_tracer_adjsens/ |
402 |
|
- add extra sub-verification which uses "SOM" advection scheme |
403 |
|
(2nd Order-Moment Advection Scheme, Prather Limiter) with |
404 |
|
#define GAD_ALLOW_TS_SOM_ADV in GAD_OPTIONS.h |
405 |
|
and in data: |
406 |
|
multiDimAdvection=.TRUE., |
407 |
|
tempAdvScheme=81, |
408 |
|
saltAdvScheme=81, |
409 |
|
o model/src: |
410 |
|
- introduce ALLOW_BALANCE_RELAX which allow the removal |
411 |
|
of the global mean of relaxation terms by setting |
412 |
|
balanceThetaClimRelax and balanceSaltClimRelax |
413 |
|
- disable balanceEmPmR and balanceQnet in the case when useSeaice. |
414 |
|
This case is now treated appropriately in seaice_growth.F |
415 |
|
o pkg/seaice: |
416 |
|
- do balanceEmPmR and balanceQnet based on the net fluxes |
417 |
|
(SItflux) that enter the ocean+ice system rather than QNET |
418 |
|
and EMPMR. Works best in real fresh water. |
419 |
|
- SIatmFW, SIatmQnet, and SItflux computations had to be moved |
420 |
|
out of the diagnotics bloc, and stored in global arrays. |
421 |
|
|
422 |
|
checkpoint63o (2012/06/29) |
423 |
|
o vertical mixing schemes: |
424 |
|
- remove unused S/R CALC_BUOYANCY (left from branch-atmos merging) |
425 |
|
- replace computation of density (in Function STATE1, opps_calc.F) by a call |
426 |
|
to S/R FIND_RHO_SCALAR |
427 |
|
- pass sigmaR to S/R PP81_CALC, MY82_CALC (but not yet used) and to GGL90_CALC |
428 |
|
and use it for N^2 calculation in ggl90_calc.F (save 2 density computations); |
429 |
|
affects output (vermix.ggl90) at truncation level. |
430 |
|
o testreport: |
431 |
|
- change adjustment of SIZE.h for mpi+mth testing: preserve enough tiles |
432 |
|
(nSx,nSy) for the number of threads (from eedata.mth) |
433 |
|
o model/src: |
434 |
|
- remove 2nd set of calls at the top of initialise_varia.F (Adjoint part) |
435 |
|
related to Non_Lin Free-Surf or Bottom-Control (but not longer necessary). |
436 |
|
o verification/tutorial_tracer_adjsens/ |
437 |
|
- compile with ALLOW_CG2D_NSA |
438 |
|
- run with nonlinFreeSurf=3 |
439 |
|
- update reference output_adm.txt |
440 |
|
o tools/genmake2: |
441 |
|
- add "-DHAVE_NETCDF" to the DEFINES list when NetCDF is available ; |
442 |
|
refine usePROFILES switching off (only when NetCDF is not available). |
443 |
|
o verification (atmospheric set-up): |
444 |
|
- change eosType='IDEALGAS' to eosType='IDEALG' to match variable declaration |
445 |
|
o pkg/profiles : refine profiles and files handling. |
446 |
|
o model/src and pkg/ecco : introduce run-time switch usePROFILES. |
447 |
|
o model/src/the_main_loop.F, forward_step.F and pkg/exf/EXF_OPTIONS.h : |
448 |
|
- changes to use model/src/the_main_loop.F rather than the ecco version |
449 |
|
in ecco type adjoint runs. |
450 |
|
o model/src: |
451 |
|
- move computation of recip of grid distance & area from INI_MASKS_ETC to INI_GRID |
452 |
|
- move setting of h0Fac[C,W,S] from INI_LINEAR_PHISURF to INI_MASKS_ETC |
453 |
|
- update_masks_etc.F: rename SMOOTH*_R4,R8 function to the corresponding type (RS,RL) |
454 |
|
o model/src: |
455 |
|
- fix S/R UPDATE_CG2D for ALLOW_DEPTH_CONTROL case (was empty before); |
456 |
|
update output of test experiment bottom_ctrl_5x5. |
457 |
|
o pkg/generic_advdiff: |
458 |
|
- implement vertical bi-harmonic diffusion (for now, only for T & S and |
459 |
|
only explicitly). |
460 |
|
o verification/bottom_ctrl_5x5/ |
461 |
|
update reference output (results changed quite some time ago, |
462 |
|
not sure when; but new reference results look ok in terms of grdchk) |
463 |
|
o verification/1D_ocean_ice_column/ |
464 |
|
For verification purposes, the exact adjoint rather than an approx. |
465 |
|
should be used (e.g., to comare against F.D. or TLM). |
466 |
|
removed autodiff_inadmode*_ad.F routines in code_ad/ |
467 |
|
Update results/ accordingly. |
468 |
|
o pkg/cal: |
469 |
|
- change cal_getdate.F so that output date is function of input time only |
470 |
|
(less ambiguous, not mixed up with input iteration number). |
471 |
|
- fix calendarDumps for non-zero phase-lag (S/R CAL_TIME2DUMP). |
472 |
|
o pkg/diagnostics: |
473 |
|
- implement calendarDumps also for stats-diags |
474 |
|
- implement calendarDumps also for snap-shot output (negative freq) |
475 |
|
- for now, ignores calendarDumps if non-zero phase-lag. |
476 |
|
|
477 |
|
checkpoint63n (2012/06/04) |
478 |
|
o model/src: |
479 |
|
- recompute rSurfW & S to ensure consistent column thickness (needed for rStar) |
480 |
|
between Sum_k(hFac*drF) and rSurf-rLow, for now only when using pkg/shelfice; |
481 |
|
- use h0FacC, W & S to compute hFac_surfC, W & S in calc_surf_dr.F (more |
482 |
|
similar to calc_r_star.F); update results of experiment internal_wave and |
483 |
|
tutorial_global_oce_in_p (affected @ truncation level). |
484 |
|
o pkg/shelfice: |
485 |
|
- add parameterisation of advective-diffusive flux into the ice shelf |
486 |
|
according to Holland and Jenkins (1999), eq.22-33, as suggested by Jan De Ryd |
487 |
|
- simply eps3/3a computation |
488 |
|
o eesupp: |
489 |
|
- add new CPP option (DISCONNECTED_TILES) to disconnect tiles (no exchange |
490 |
|
between tiles, just fill-in edges assuming locally periodic subdomain) and |
491 |
|
add corresponding exchange subroutines (EXCH0); |
492 |
|
- new CG-solver version (_EX0) for disconnected-tiles special case. |
493 |
|
o model/src (CG-solvers): |
494 |
|
- modify Single-Reduction version (CG2D_SR) in case cg2dMaxIters is reached |
495 |
|
(similar final residual output and max-Iter number as with standard version); |
496 |
|
- allows to store and use lowest-residual solution (standard & _SR versions); |
497 |
|
- cleaning up (remove old, commented out code, LETS_MAKE_JAM ...). |
498 |
|
- by default, also compile CG2D_SR version (define ALLOW_SRCG in CPP_OPTIONS.h) |
499 |
|
o eesup/src: |
500 |
|
- global_vec_sum.F cleaning: add a BARRIER ; update comments ; remove ndef |
501 |
|
ALWAYS_USE_MPI (no longer needed after fixing usingMPI settings - March 2012) |
502 |
|
o pkg/salt_plume added to lab_sea/input_ad.noseaicedyn |
503 |
|
|
504 |
|
checkpoint63m (2012/04/27) |
505 |
|
o pkg/shelfice: |
506 |
|
- account for partical cells in surface level below ice-shelf for hydrostatic |
507 |
|
pressure calculion (set uniformFreeSurfLev to F); update isomip experiments. |
508 |
|
o ad test experiments: |
509 |
|
- update and rename to "ad_optfile.local" local AD-optfile specific to a |
510 |
|
particular adjoint verification experiment (in verification/*/code_ad/) |
511 |
|
o pkg/seaice & pkg/exf: |
512 |
|
- Part two of ECCO-CLIVAR code merges. |
513 |
|
Completes merge of seaice area relax. code. |
514 |
|
Requires unified CPP option EXF_ALLOW_SEAICE_RELAX |
515 |
|
to be set in EXF_OPTIONS.h since this pkg controls its I/O |
516 |
|
(and avoids second CPP option in SEAICE_OPTIONS.h) |
517 |
|
- change default value of costIceStart & costIceEnd to a valid date. |
518 |
|
- add snowPrecip to HSNOW in seaice_growth |
519 |
|
o pkg/ptracers: |
520 |
|
- move ptracers summary and setup check into separated S/R ; add a check for |
521 |
|
valid PTRACERS_Iter0 value (error+stop or just warning). |
522 |
|
o pkg/cal (+ pkg/ecco): |
523 |
|
- avoid calling a S/R with 2 times the same arg (FWD: unless both are only used |
524 |
|
as input; and for AD, unless both are not differentiable (e.g., k index)). |
525 |
|
o model/src: |
526 |
|
- define internal flag "uniformFreeSurfLev" (=T when free-surface level-index |
527 |
|
is everywhere k=1) to speed-up some calculations (by pass kSurf reference); |
528 |
|
temporary set to usingZCoords value (to be changed to F when using shelfice). |
529 |
|
- hydrostatic pressure: implement partial cell near the surface also for |
530 |
|
Z-coordinates (for the case where uniformFreeSurfLev=F); replace test on |
531 |
|
usingZCoords with test on uniformFreeSurfLev where needed (in NonHyd code). |
532 |
|
o pkg/ctrl/ctrl_init.F: |
533 |
|
- add a formal parameter to s/r ctrl_init_rec to print |
534 |
|
xx_$(ctrl_valiable)_file for easier debugging |
535 |
|
- move call of cal_* routines into the if-block where they are really |
536 |
|
needed so that xx_$(ctrl_variable)_period can be 0 or -12 without having |
537 |
|
to specify a startdate |
538 |
|
o pkg/seaice: |
539 |
|
- update comments in SEAICE_OPTIONS.h (+ similar update of lab_sea/code and |
540 |
|
global_ocean.cs32x15/code versions). |
541 |
|
- turn on seaice Multi-category (in data.seaice, SEAICE_multDim=7) for |
542 |
|
verification experiment global_ocean.cs32x15.seaice and update results. |
543 |
|
o pkg/exf: |
544 |
|
- new S/R EXF_GETFIELD_START to return starting-time of a forcing field |
545 |
|
(according to useExfYearlyFields). |
546 |
|
o pkg/cal: |
547 |
|
- move calendar setting & summary (previously in cal_readparams.F) into new |
548 |
|
S/R CAL_INIT_FIXED to be called later in the initialisation process, once |
549 |
|
length of the simulation is set (after SET_PARMS call); This is an attempt |
550 |
|
to fix Pb when, e.g.: pkg/longstep, modifies the length of the run. |
551 |
|
- For safety: keep a record of pkg/cal parameter settings status |
552 |
|
to allow to STOP if one pkg/cal S/R is called too early. |
553 |
|
- add a stop in all basic pkg/cal subroutines if called before setting is done. |
554 |
|
- modify cal_getdate.F: stop when nothing was returned; |
555 |
|
if called with myIter=-1, simply retrun modelBaseDate (1,2 = startDate_1,_2); |
556 |
|
this allows to remove "cal.h" from seaice_read_parms.F & init_mnc_vars.F. |
557 |
|
|
558 |
|
checkpoint63l (2012/04/04) |
559 |
|
o pkg/bbl: |
560 |
|
- fix call to FIND_RHO_2D in bbl_calc_rho.F (case multi-tiles per proc); |
561 |
|
update global_with_exf output.yearly.txt file. |
562 |
|
o pkg/cal: |
563 |
|
- start to clean-up cal_set.F ; fix for the case where baseTime is non-zero. |
564 |
|
- remove 2 unsused and very short (1 line only) functions ; |
565 |
|
- remove unsused subroutines from AD-Diff list. |
566 |
|
o model/src: |
567 |
|
- start to implement clean wait to stop in initialization S/R (config_check.F, |
568 |
|
packages_check.F, ini_parms.F and pkg/obcs/obcs_check.F + pkg/mypackage) |
569 |
|
o folowing PRINT_MESSAGE changes: |
570 |
|
- remove reminder multiple print to the same file (STDOUT) when multi-threads. |
571 |
|
o pkg/bbl: |
572 |
|
- modification so it can work with z* |
573 |
|
- replace deltaT with dTtracerLev(k) for asynchronous time stepping |
574 |
|
- global_with_exf/input.yearly verification; changes output.yearly.txt |
575 |
|
o eesupp/src: |
576 |
|
- PRINT_MESSAGE: use OpenMP critical instruction (similar to PRINT_ERROR); |
577 |
|
writing to unit 0 (if unit=errorMessageUnit) extended to the case myThid=1 |
578 |
|
- CHECK_THREADS: specific code if using OpenMP threading: improve checking of |
579 |
|
# of threads across processors (e.g., different env OMP_NUM_THREADS setting) |
580 |
|
- improve special preliminary call to EESET_PARMS (if eedata is read 2 times). |
581 |
|
- when using MPI, always set usingMPI=T ; set default accordingly and stop |
582 |
|
if not consistent. |
583 |
|
- ALL_PROC_DIE: add a BARRIER after MPI_FINALIZE call ; skip this when thread |
584 |
|
argument is zero (in case called within single-thread section). |
585 |
|
o pkg/fizhi: |
586 |
|
clean-up turbulence cold-start switch: decided in fizhi_init_vars.F, stored |
587 |
|
in common bloc (fizhi_coms.h) and then passed as argument up to S/R TURBIO. |
588 |
|
o pkg/exch2: |
589 |
|
- add argument "updateCorners" to S/R exch2_get_uv_bounds (enable to |
590 |
|
activate EXCH_IGNORE_CORNERS in vector EXCH S/R). |
591 |
|
- switch 1rst EXCH call to IGNORE_CORNERS (instead of UPDATE_CORNERS) |
592 |
|
but keep the 2nd with UPDATE_CORNERS: this prevents overwriting good data |
593 |
|
with bad ones coming from a not-yet-updated halo. |
594 |
|
This should fix the problem when sNx <> sNy and GCD(sNx,sNy) < 2*Olx. |
595 |
|
o pkg/frazil: |
596 |
|
move FrzForcT diag from frazil_diagnostics_state.F to frazil_calc_rhs.F |
597 |
|
http://mitgcm.org/pipermail/mitgcm-devel/2012-March/005384.html |
598 |
|
o model/src/config_check.F: add a check for OLx>sNx/OLy>sNy |
599 |
|
o verification/lab_sea |
600 |
|
* #undef SEAICE_EXCLUDE_FOR_EXACT_AD_TESTING |
601 |
|
* #define ALLOW_SST_CONTROL & ALLOW_SST_CONTROL |
602 |
|
* nTimeSteps 3 -> 4 |
603 |
|
* changed reference output accordingly |
604 |
|
o pkg/ctrl: |
605 |
|
* fix ctrl_get_gen for xx_sst, xx_sss |
606 |
|
o pkg/fizhi: |
607 |
|
- fix 2 calls to S/R GRD2MSC (argument igrd) in update_earth_exports.F |
608 |
|
update results of exp. fizhi-cs-32x32x40 (fail with only 6 digits for cg2d) |
609 |
|
- by default, use new version of S/R GETPWHERE (#define TRY_NEW_GETPWHERE); |
610 |
|
fix few out-of-bounds indices in referencing vars (initialise igrd to 1); |
611 |
|
clean-up fizhi_turb.F (remove unused & uninitialised qdiaglsm array). |
612 |
|
o pkg/gmredi: |
613 |
|
- add diagnostic of Redi (main diagnonal) vertical diffusive flux of Temp |
614 |
|
o pkg/fizhi: |
615 |
|
- fix few initialisation and multi-tile Pb: aqualev & gridalt now pass the |
616 |
|
(old)aces open64 tuv test. |
617 |
|
o pkg/icefront: fix bug and reorder loop in icefront_init_varia.F |
618 |
|
o pkg/mom_vecinv: |
619 |
|
- fix 1 loop range in mom_vecinv.F to be like all the other loops which |
620 |
|
set/update gu,gv/Diss (fix Pb of uninitialised fVerUkp in admom_vecinv |
621 |
|
with non-lin free-surf). |
622 |
|
o model/src: move k-loop outside of i/j-loops when (re-)computing R_low |
623 |
|
in ini/update_masks_etc.F in order to avoid -O3 optimization problems with |
624 |
|
some compilers (gfortran 4.6.0 on MacOS) |
625 |
|
o momentum: |
626 |
|
- separate fVer?(:,:,kUp) & fVer?(:,:,kDown) in argument list of MOM_FLUXFORM |
627 |
|
& MOM_VECINV subroutines (to help TAF). This fixes Pb of uninitialised |
628 |
|
fVer?(kUp) in admom_fluxform with non-lin free-surf or bottom control. |
629 |
|
|
630 |
|
checkpoint63k (2012/03/17) |
631 |
|
o model/src: |
632 |
|
- selectAddFluid (ALLOW_ADDFLUID): add missing term in horizontal momentum |
633 |
|
flux-form (mom_fluxform.F) and vertical momentum (calc_gw.F) equations. |
634 |
|
- SOLVE_DIAGONAL: switch default to use adjoinable & vectorized version |
635 |
|
(#undef SOLVE_DIAGONAL_LOWMEMORY); update output of exp. front_relax.bvp. |
636 |
|
o model/src: solve_*diagonal.F |
637 |
|
- replace division with multiplication by inverse (similar to original version) |
638 |
|
update output of AD exp. global_ocean.90x40x15. |
639 |
|
- fix pkg/gmredi/gmredi_calc_psi_bvp.F & pkg/ggl90/ggl90_calc.F |
640 |
|
to enable the use of the vectorized & differentiable solve_tridiagonal.F |
641 |
|
- test exp. global_ocean.90x40x15/input_ad: switch to advection scheme 20 for |
642 |
|
vertical advection (to get solve_tridiagonal tested in AD mode). |
643 |
|
o pkg/seaice: ECCO-CLIVAR merge: |
644 |
|
1. seaice AREA relaxation |
645 |
|
* CPP option: SEAICE_ALLOW_AREA_RELAXATION |
646 |
|
* runtime flag: SEAICE_tauAreaObsRelax |
647 |
|
2. modified global mean cost imbalance code: CPP option |
648 |
|
ALLOW_SSH_GLOBMEAN_COST_CONTRIBUTION |
649 |
|
o another iteration on solve_*diagonal.F |
650 |
|
Make choices independent of AUTODIFF |
651 |
|
Default will be the adjoinable vectorized version. |
652 |
|
Alternative choices are: |
653 |
|
* non-adjoinable original version with low memory footprint: |
654 |
|
#define SOLVE_DIAGONAL_LOWMEMORY |
655 |
|
* still adjoinable but does not vectorize (k as inner-most loop): |
656 |
|
#define SOLVE_DIAGONAL_KINNER (keeping SOLVE_DIAGONAL_LOWMEMORY undef) |
657 |
|
o testreport |
658 |
|
- add option to use a home made wrapper around "make" |
659 |
|
- remove MITGCM_IEEE env. variable (no longer active) |
660 |
|
o pkg/shelfice |
661 |
|
- improve computation of friction velocity for velocity dependent |
662 |
|
exchange coefficients: include velocities into SHELFICEBoundaryLayer |
663 |
|
block to average velocities vertically |
664 |
|
- add TAF store directives for SHI_ALLOW_GAMMAFRICT |
665 |
|
o Modify diagonal solvers, now three versions (upon request): |
666 |
|
#ifndef SOLVE_DIAGONAL_AUTODIFF |
667 |
|
1. default version (dating back a while ago) |
668 |
|
#else /* def SOLVE_DIAGONAL_AUTODIFF */ |
669 |
|
# ifdef SOLVE_DIAGONAL_KINNER |
670 |
|
2. adjoinable version with k-innermost loop (G. Forget) |
671 |
|
# else |
672 |
|
3. adjoinable vectorized version of 2. |
673 |
|
# endif |
674 |
|
#endif /* SOLVE_DIAGONAL_AUTODIFF */ |
675 |
|
All flags are currently set within routines, but could/should be moved to CPP-options. |
676 |
|
Array bounds for 1. are as before, for 2.,3. cover the entire domain |
677 |
|
(could be changed for 1. too). |
678 |
|
o verification/lab_sea (fwd): |
679 |
|
- update the 4 fwd lab_sea test experiment: no longer SEAICE_GROWTH_LEGACY, |
680 |
|
use 7 categories (except hb87), new solve4temp params (except salt_plume), |
681 |
|
constant salinity (=4) or variable salinity (salt_plume) done through |
682 |
|
SItracer. |
683 |
|
o pkg/seaice: |
684 |
|
- rename parameters: SIsalFrac to SEAICE_saltFrac & SIsal0 to SEAICE_salt0; |
685 |
|
- remove old parameters ( SEAICE_freeze, SEAICEturbFluxFormula, SEAICE_gamma_t, |
686 |
|
SEAICE_gamma_t_frz, SEAICE_availHeatTaper & SEAICE_availHeatFracFrz) |
687 |
|
from SEAICE_PARMS.h ; still need to fix seaice_cost_sst.F (broken when |
688 |
|
SEAICE_freeze got retired) but already better than previously with |
689 |
|
SEAICE_freeze=123456.7 deg.C; |
690 |
|
- move setting of facOpenGrow/Melt from SEAICE_INIT_FIXED to SEAICE_READPARMS |
691 |
|
(safer multi-threaded setting); always set SEAICEuseEVP; |
692 |
|
- setting of ocean-ice turb. flux coeff: moved from SEAICE_CHECK to |
693 |
|
SEAICE_READPARMS & SEAICE_INIT_FIXED (this fixes wrong summary report); |
694 |
|
stop if multiple specifications for the same coeff; make sure default |
695 |
|
SEAICE_mcPheePiston is compatible with drF(1) & deltaT. |
696 |
|
o pkg/obcs: |
697 |
|
- allow to switch to upwind 1rst order advection scheme for the advective |
698 |
|
flux computation at the open-boundary; disabled for adjoint built. |
699 |
|
o pkg/ptracers: |
700 |
|
- rename PTRACERS_RESTART.h -> PTRACERS_START.h |
701 |
|
- rename ptracers_fields_blocking_exchanges.F -> ptracers_fields_blocking_exch.F |
702 |
|
- add parameter to decide when to start a tracer; use internal flag to |
703 |
|
know which tracer to step-forward; switch this internal flag at the top |
704 |
|
of forward_step.F (new S/R PTRACERS_SWITCH_ONOFF); for now, not enabled |
705 |
|
for adjoint built. |
706 |
|
o pkg/seaice: |
707 |
|
- add SEAICE_SIZE.h anywhere in the code where SEAICE.h is included ; |
708 |
|
- move MULTDIM declaration and setting from SEAICE.h to SEAICE_SIZE.h ; |
709 |
|
- keep experiment offline_exf_seaice truly single category (set MULTDIM=1). |
710 |
|
o pkg/ecco: |
711 |
|
- fix writing of output file DiagnosCost_ClimSalt/Theta |
712 |
|
and DiagnosCost_DriftSalt/Theta (if multi-tiles per proc). |
713 |
|
o pkg/seaice: |
714 |
|
- add missing initialisation of COR_ICE array in dynsolver.F |
715 |
|
- merging SEAICE_MULTICATEGORY define and undef cases. Now the same code is |
716 |
|
compiled either way, and is tested in all exps (multidim was only tested |
717 |
|
in lab_sea ad before). |
718 |
|
- cosmetic change to prepare for future addition of ITD. no change to results. |
719 |
|
no defaults change. |
720 |
|
- added run time param SEAICE_multDim (7 for SEAICE_MULTICATEGORY, 1 otherwise). |
721 |
|
- by lack of adequate verification experiments, I further used custom versions |
722 |
|
of global_ocean.cs32x15 to test multicat or not, with pickup or not. All |
723 |
|
should be correct. |
724 |
|
- avoid recomputations in ad. |
725 |
|
o array initialisation in adjoint mode: |
726 |
|
1) pkg/mom_common/mom_calc_ke.F: |
727 |
|
add full array initialisation of KE inside mom_calc_ke.F (ALLOW_AUTODIFF_TAMC): |
728 |
|
The S/R which calls MOM_CALC_KE initialises KE but in some recomputation |
729 |
|
part this initialisation is gone, resulting in Floating Point Exception |
730 |
|
(caught by open64 compiler with debug option). |
731 |
|
2) model/src/convective_weights.F |
732 |
|
set output convectCount over the full array range (instead of doing |
733 |
|
an update): this fix a similar issue when called within recomputation (adjoint) |
734 |
|
which could by-pass the initialisation (normally done in the caller S/R). |
735 |
|
o pkg/seaice: |
736 |
|
- final version of turbulent flux params, as agreed at MIT ice meet: |
737 |
|
SEAICE_mcPheePiston (default is ~ 0.000875) |
738 |
|
SEAICE_mcPheeTaper (default is 0.) |
739 |
|
SEAICE_frazilFrac (defualt is 1.) |
740 |
|
SEAICE_mcPheeStepFunc (default is false) |
741 |
|
- changes results at trunction error level |
742 |
|
- ifdef SEAICE_EXTERNAL_FLUXES, disallow SEAICE_waterAlbedo |
743 |
|
o added pkg/frazil: |
744 |
|
- see frazil_description.tex for details |
745 |
|
- verification experiment is global_with_exf/input.yearly; changes output.yearly.txt |
746 |
|
o pkg/shelfice/shelfice_thermodynamics.F |
747 |
|
- fix exponents for velocity dependent exchange coefficients: change |
748 |
|
from 2/3(=0) to 0.66666666666666667 |
749 |
|
- move computation of velocity dependent exchange coefficients into |
750 |
|
separate loop, so that it applies to both thermodynamics |
751 |
|
- move some divisions by constants out of loops |
752 |
|
|
753 |
|
checkpoint63j (2012/02/16) |
754 |
|
o pkg/seaice: |
755 |
|
- overhaul of seaice_summary.F after feb 2012 clean-up of seaice_growth.F etc. |
756 |
|
- SItracer: added run time params, and allow coupling to ocn salinity. |
757 |
|
o pkg/seaice: remove (almost) all divisions by constants within loops |
758 |
|
in seaice_growth.F and seaice_solve4temp.F, this changes (only) the cg2d |
759 |
|
results of global_ocean.cs32x15.seaice from 16 to 11 digits (but only |
760 |
|
those, caused by 1/QI -> recip_QI) -> update output.seaice.txt, |
761 |
|
output_adm.seaice_dynmix.txt and also lab_sea: output_adm.evp.txt |
762 |
|
o pkg/seaice : |
763 |
|
- simplify McPhee tapering, and allow to use it with SEAICEturbFluxFormula |
764 |
|
= 1 or 2, using newly added run time parameter SEAICE_availHeatTaper. |
765 |
|
- fix d_AREAbyATM, d_AREAbyOCN, d_AREAbyICE diags I broke in r1.148. |
766 |
|
- add permanent SItflux diag that corresponds to TFLUX but includes |
767 |
|
ice+snow. Hence SItflux-TFLUX should match the ice+snow heat budget. |
768 |
|
- allow activation/testing of a fix for suspected missing term in |
769 |
|
ocn-ice heat budget (to be confirmed). To test this, you want to |
770 |
|
undef SEAICE_DISABLE_HEATCONSFIX, and then set the run time param |
771 |
|
SEAICEheatConsFix to .TRUE. that also allows the 'SIaaflux' diagnostic. |
772 |
|
- no change in results, fwd or ad. |
773 |
|
o global_ocean.cs32x15/results: 3 changes following from the pkg/seaice commit below: |
774 |
|
- switch to variable freezing point. |
775 |
|
Results change : cg2d -> 1 digits; ad -> 3-6-6 digits. |
776 |
|
- bug fix : the sublimation term that was missing is now activated. |
777 |
|
Results change : cg2d -> 2 digits; ad -> 8-6-6 digits. |
778 |
|
- bug fix : flooding sign term for SIsal0.NE.0. |
779 |
|
Result change : cg2d -> 8 digits; ad -> 11-12-9 digits. |
780 |
|
o pkg/seaice: |
781 |
|
- unified freezing point treatment : the old SEAICE_VARIABLE_FREEZING_POINT |
782 |
|
is now the default, and the old default constant freezing point is recovered |
783 |
|
with SEAICE_tempFrz0 = -1.96, |
784 |
|
SEAICE_dTempFrz_dS = 0., |
785 |
|
- retire SEAICE_freeze (= the old way of specifycing the constant freezing point) |
786 |
|
- remove ALLOW_SEAICE_FLOODING brackets; run time switch is already there. |
787 |
|
- bug fix (thanks to M. Losch) : the sublimation term that was missing is now |
788 |
|
activated. To reproduce old results with this bug define SEAICE_DISABLE_SUBLIM. |
789 |
|
- bug fix (silly me) : flooding sign term (seaice_growth.F r 1.149line 1211) |
790 |
|
was wrong for SIsal0.NE.0. Changes cs32x15 results (see above item for detail). |
791 |
|
- bug fix (thanks to O. Jahn) : area loss for melting got messed up in |
792 |
|
seaice_growth r149 for legacy branch (no results change). |
793 |
|
- introduce SEAICE_CAP_SUBLIM : caps sublimation heat flux in solve4temp (code |
794 |
|
from I. Fenty). |
795 |
|
- results did not change except for global_ocean.cs32x15, mostly due to the |
796 |
|
switch to variable freezing point (see above item for detail). |
797 |
|
o pkg/thsice: |
798 |
|
- fix and simplify interface to salt_plume pkg ; diagnose directly amount of |
799 |
|
seawater freezing in thsice_calc_thickn (not used). |
800 |
|
o pkg/seaice: turbulent flux term specification via run time params. |
801 |
|
- replace CPP switches with run time switches to facilitate testing |
802 |
|
- retired CPPs : MCPHEE_OCEAN_ICE_HEAT_FLUX, |
803 |
|
GRADIENT_MIXED_LAYER_TURBULENCE_FACTOR |
804 |
|
- added runtime parameter : SEAICEturbFluxFormula |
805 |
|
- move Mcphee constants to SEAICE_PARAMS.h |
806 |
|
- seaice_readparms.F : read turbulent flux term params (SEAICEturbFluxFormula; |
807 |
|
SEAICE_availHeatFrac/Frz, or SEAICE_gamma_t/_frz if apply); |
808 |
|
remove seaice_summary call (moved to seaice_init_fixed) |
809 |
|
- seaice_init_fixed.F : set SEAICE_availHeatFrac/Frz depending on |
810 |
|
SEAICEturbFluxFormula, and seaice_summary call (moved from seaice_readparms) |
811 |
|
- seaice_summary.F : add missing report of read turbulent flux term set up |
812 |
|
and SITRACER. |
813 |
|
- seaice_check.F : check range; inform and stop if retired CPPs are used |
814 |
|
(incl. SEAICE_AGE). |
815 |
|
- seaice_growth.F : unified use of SEAICE_availHeatFrac/Frz, without |
816 |
|
CPP switch, same computation as before for the different options. |
817 |
|
- change results at the truncation level. |
818 |
|
o pkg/seaice: |
819 |
|
- remove un-used parameters LAD & MAX_TICE ; add new run-time parameters: |
820 |
|
useMaykutSatVapPoly (default=F), postSolvTempIter (default=2) and |
821 |
|
SEAICE_wetAlbTemp (default=-1e-3). Remove option SEAICE_SOLVE4TEMP_LEGACY. |
822 |
|
- update verification/*/input*/data.seaice: |
823 |
|
to reproduce previous results with former #defined SEAICE_SOLVE4TEMP_LEGACY |
824 |
|
code, set useMaykutSatVapPoly=T & postSolvTempIter=0 ; |
825 |
|
and to reproduce previous results with former non-SOLVE4TEMP_LEGACY code, |
826 |
|
set SEAICE_wetAlbTemp=0 & SEAICE_snowThick=0. |
827 |
|
o pkg/seaice/seaice_solve4temp.F (SOLVE4TEMP_LEGACY): |
828 |
|
- remove LEGACY code for solving for tsurf (A1,A2,A3) but maintain the same |
829 |
|
algorithm (same choice: useMaykutPolySatVap=T, postSolvTempIter=0); |
830 |
|
difference in results only due to machine truncation. |
831 |
|
- remove MAX_TICE (tsurf is always =< TMELT anyway); keep MIN_TICE if using |
832 |
|
MaykutPolySatVap; keep MIN_LWDOWN. |
833 |
|
- adapt SEAICE_MODIFY_GROWTH_ADJ code (untested) to non-legacy formulation. |
834 |
|
- update output of verification exp. lab_sea (4 fwd: 11 cg2d digits except hb87, |
835 |
|
+ 3 AD: 6 digits except evp) and offline_exf_seaice.seaice (12 digits). |
836 |
|
o pkg/salt_plume: |
837 |
|
- isolate saltPlumeFlux array exch into a specific S/R called from |
838 |
|
do_oceanic_phys.F |
839 |
|
- add interface to pkg/salt_plume (filling of saltPlumeFlux) in pkg/thsice. |
840 |
|
o verification: |
841 |
|
- move untested set-up to MITgcm_contrib/verification_other. |
842 |
|
o pkg/seaice: |
843 |
|
- retire old way of SEAICE_AGE. Now replaced with ALLOW_SITRACER. |
844 |
|
o pkg/seaice: |
845 |
|
- seaice_solve4temp.F (keeping results unchanged): |
846 |
|
* add 2 future run-time params to select which saturation vapor pressure |
847 |
|
formula is used and how post-iteration fluxes are set. |
848 |
|
Temporary set these 2 params according to option SOLVE4TEMP_LEGACY; |
849 |
|
* New piece of code (not used) to update surf. atmos. heat flux F_ia using |
850 |
|
the linearized expression to stay consistent with tsurf finding. |
851 |
|
- seaice_freedrift.F avoid floating exception in sqrt(y) since y=sqrt(x*x)-x |
852 |
|
can be < 0. |
853 |
|
o pkg/seaice/seaice_solve4temp.F (SOLVE4TEMP_LEGACY undef): |
854 |
|
- remove from F_ia (and from it's derivative dFiDTs1) the contribution of |
855 |
|
conductive heat flux F_c , and add it explicitly when updating tsurf as |
856 |
|
solution of: Fc = Fia + d/dT(Fia - Fc) *delta.tsurf |
857 |
|
so that now F_ia has a consistent meaning through the entire routine. |
858 |
|
This causes differences at truncation level: update output of fwd exp. |
859 |
|
global_ocean.cs32x15.seaice. |
860 |
|
o pkg/seaice: |
861 |
|
- lots of editing/cleaning in seaice_solve4temp.F + better documentation |
862 |
|
of arrays, sign/direction of fluxes; no effect on results. |
863 |
|
- replace hard coded 273.16 (=previous value of celsius2K) by celsius2K |
864 |
|
in seaice_budget_ocean.F & seaice_solve4temp.F (SOLVE4TEMP_LEGACY part); |
865 |
|
note: this should have been done before changing celsius2K default. |
866 |
|
- to keep results unchanged, add back old value of celsius2K in verification |
867 |
|
exp. input*/data : lab_sea (AD), offline_exf_seaice (fwd+AD) & seaice_obcs |
868 |
|
- replace hard coded seawater freezing temp in seaice_solve4temp.F (part |
869 |
|
SOLVE4TEMP_LEGACY) of 271.20 (=previous value of celsius2K + default |
870 |
|
value of SEAICE_freeze) by celsius2K+SEAICE_freeze so that seawater |
871 |
|
freezing temp is consistent with value in seaice_growth.F ; this causes |
872 |
|
differences at machine truncation level: update lab_sea AD output (3/4). |
873 |
|
o pkg/seaice: |
874 |
|
- mask (in exf_init_fixed.F) atemp,aqh,evap,precip,runoff,lwflx,areamask |
875 |
|
and climsst/sss even when using seaice (to prevent calculation that might |
876 |
|
use SST on land); does not affect any verification exp. |
877 |
|
- apply mask to taux,tauy (in seaice_get_dynforcing.F); update output |
878 |
|
of lab_sea.fd & lab_sea.hb87 and global_ocean.cs32x15.seaice_dynmix. |
879 |
|
o pkg/seaice/seaice_growth.F: add mask to avoid non-zero ice thickness |
880 |
|
over land (non-legacy code) |
881 |
|
o vectorize salt_plume package: |
882 |
|
- add extra code to salt_plume_frac and salt_plume_tendency_apply_s |
883 |
|
- this code computes plumebk everywhere (also over land), thus requires |
884 |
|
code to avoid divisions by zero and catch the case of kLev=Nr. |
885 |
|
- efficiency of salt_plume_frac is still sub-optimal |
886 |
|
o tools (genmake2): |
887 |
|
- fix makefile NAME specification (option: -makefile NAME) by adding |
888 |
|
"-f $MAKEFILE" in few places where it was missing including makedepend |
889 |
|
command (contribution from Paul). |
890 |
|
|
891 |
|
checkpoint63i (2012/01/24) |
892 |
|
o model/src: |
893 |
|
- add factorized versions of some EOS code to find_rho.F, |
894 |
|
define USE_FACTORIZED_EOS to turn it on (default for TARGET_NEC_SX) |
895 |
|
o model/src: |
896 |
|
- add celsius2K to namelist PARM01; remove unused recip_rhoNil from PARAMS.h; |
897 |
|
- change default celsius2K from 273.16 to 273.15 |
898 |
|
- to keep results unchanged, add back old value of celsius2K in verification |
899 |
|
exp. input*/data : aim.5l_cs, 1D_ocean_ice_column, lab_sea (fwd), |
900 |
|
global_ocean.cs32x15(seaice) & cpl_aim+ocn(input_atm). |
901 |
|
o model/src: add a little initialisation trick to find_rho.F and find_alpha.F |
902 |
|
to make TAF generate vectorizable code |
903 |
|
o pkg/seaice/seaice_evp.F: add code (copied from CICE) that avoids underflows. |
904 |
|
Turn on by defining SEAICE_EVP_ELIMINATE_UNDERFLOWS. Most compilers |
905 |
|
have flags that do this more efficiently. |
906 |
|
o pkg/diagnostics/diagstats_calc.F |
907 |
|
- for TARGET_NEC_SX fix the treatment of the scaling factor tmpFac=scaleFact |
908 |
|
o pkg/obcs (with exch2): |
909 |
|
- derive OB indices in overlap region from OB-InsideMask (in obcs_init_fixed.F) |
910 |
|
and remove code (in obcs_readparms.F) which was based on EXCH of indices |
911 |
|
(through buffer) and did not account for exch2 (e.g., CS-grid) topology. |
912 |
|
o pkg/exf and seaice exp: |
913 |
|
- change offline_exf_seaice.seaicetd test exp. to use default DIFF1 value (=0) |
914 |
|
so that switching off area & heff advection flags does not affect results. |
915 |
|
- change default EXF_LWDOWN_WITH_EMISSIVITY to #define in EXF_OPTIONS.h |
916 |
|
- keep exp. seaice_obcs & 1D_ocean_ice_column(ad) unchanged (using #undef |
917 |
|
EXF_LWDOWN_WITH_EMISSIVITY) and, using pkg/exf default emissivities, update |
918 |
|
results of verification experiments: 1D_ocean_ice_column(fwd), |
919 |
|
global_ocean.cs32x15.seaice(fwd+ad) & seaice_dynmix(ad), lab_sea (all 8) |
920 |
|
and offline_exf_seaice (all 4). |
921 |
|
o pkg/seaice: |
922 |
|
- remove array YNEG, RIVER, TMIX (replaced by local var) from common block; |
923 |
|
remove also areaNm1 & hEffNm1 when SEAICE_GROWTH_LEGACY is undef. |
924 |
|
- fix bug: areaNm1 & hEffNm1 are needed in seaice_growth (SEAICE_GROWTH_LEGACY) |
925 |
|
but were not filled when resp. SEAICEadvArea=F & SEAICEadvHeff=F and |
926 |
|
not multi-dim advection; |
927 |
|
- fix seaice_growth.F to compile with both SEAICE_GROWTH_LEGACY and |
928 |
|
SHORTWAVE_HEATING #undef. |
929 |
|
o optim: make Makefile more robust (?) |
930 |
|
o pkg/obcs: |
931 |
|
* fix treatment of obcs_read_pickup for TAF |
932 |
|
o pkg/shelfice: |
933 |
|
* add missing initialisation of num_shifwflx |
934 |
|
* new adjoint verification exp. for isomip |
935 |
|
* change Prandtl and Schmidt numbers from generic formulations |
936 |
|
(in terms of model diffus. and visc.) to separate runtime |
937 |
|
parameters, with Holland and Jenkins (1999) default values. |
938 |
|
o pkg/exf (interpolation): |
939 |
|
- extended input field (2 rows) near the N & S poles: |
940 |
|
* fill in with the symetric value (when even Nb of data in longitude) |
941 |
|
* add average value at the poles only for scalar quantities ; for vector |
942 |
|
component interpolation, skip the averaging and keep duplicated values, |
943 |
|
unless the 2 components have same location (uvInterp=T) in which case |
944 |
|
interpolate the 2 components together using S/R EXF_INTERP_UV. |
945 |
|
- new S/R EXF_INTERP_UV to interpolate the 2 components of a vector field |
946 |
|
together: this allows to account properly for local orientation when |
947 |
|
averaging at N & S pole. |
948 |
|
- add CPP option EXF_USE_OLD_INTERP_POLE to recover old results regarding |
949 |
|
exf-interpolation near N & S poles. Undef by default. |
950 |
|
- add debug check/print on 1rst iter if debugLevel >=2. |
951 |
|
o pkg/exf: |
952 |
|
- change masking flags definition (no longer "parameter"), put them in |
953 |
|
common block and set them in S/R EXF_INIT_FIXED. |
954 |
|
- change recognized masking flag from s,u,v to c,w,s since the mask |
955 |
|
to use (maskC,W,S) depends on position on model grid and has little to |
956 |
|
do with scalar/vector difference; fix masking if using P-coordinates. |
957 |
|
- fix masking of wind-stress (was wrong if interpolated or read on A-grid) |
958 |
|
and update results of exp. global_with_exf (x2). |
959 |
|
- fix setting of zenith-angle table when SHORTWAVE_HEATING is undef. |
960 |
|
o pkg/obcs |
961 |
|
- S/R obcs_calc_stevens: fix a small bug (gammat -> gammas where appropriate) |
962 |
|
o pkg/exf (interpolation): |
963 |
|
- assume periodicity in X only if input field cover full longitude range. |
964 |
|
- improve search for lat. index (supposed to be faster, in ~log2(ny) steps, |
965 |
|
and should vectorise). |
966 |
|
- fix input lat of the 2 added row (in case we provide N.pole data). |
967 |
|
|
968 |
|
checkpoint63h (2011/12/30) |
969 |
|
o pkg/seaice: |
970 |
|
- change declaration of UVM from _RL to _RS in order to get B-grid |
971 |
|
uIce,vIce monitor when _RS = real*4 |
972 |
|
o pkg/offline: |
973 |
|
- new S/R OFFLINE_RESET_PARMS for resetting main-model params (switch off flags); |
974 |
|
- compute vertical velocity if not read from file (switch on exactConserv |
975 |
|
& staggerTimeStep); |
976 |
|
- fix output of ptracer monitor iter & time for default PTRACERS_monitorFreq. |
977 |
|
o verification: |
978 |
|
- update grid-files (in tutorial_held_suarez_cs/input and fizhi-cs-32x32x40/input) |
979 |
|
with grid-angle generated using "utils/matlab/cs_grid/cubeCalcAngle.m": |
980 |
|
fix Pb in yG definition @ 2 missing corners (previously yG was averaged); |
981 |
|
- update output of exp. global_ocean.cs32x15.viscA4 after changing |
982 |
|
grid-file angles (used here for coriolis-3d); |
983 |
|
- remove (or gzip) few files not needed to run fizhi-cs experiments. |
984 |
|
o pkg/exf (vector interpolation): |
985 |
|
- rotate vector component toward local grid-axis orientation using grid-angles |
986 |
|
(stored in GRID.h) instead of locally computed rotation angle; keep old code |
987 |
|
available with exf option EXF_USE_OLD_VEC_ROTATION (undef by default). |
988 |
|
o pkg/seaice: |
989 |
|
- allow to compile with exf options ALLOW_ATM_TEMP or ALLOW_DOWNWARD_RADIATION |
990 |
|
undefined. Fix and remove the stop when ALLOW_RUNOFF is undefined. |
991 |
|
o pkg/PKG_OPTIONS.h files: |
992 |
|
- move #include "CPP_OPTIONS.h" just after #include "PACKAGES_CONFIG.h", |
993 |
|
outside of the #ifdef ALLOW_THISPKG / #endif block. |
994 |
|
- update verification/*/code/PKG_OPTIONS.h files. |
995 |
|
o add a verification experiment (offline_exf_seaice/input.seaice) that test |
996 |
|
the B-grid LSOR solver |
997 |
|
o pkg/seaice: sort out emissivity and long wave radiation step 2: |
998 |
|
- add emissivity for snow and ice as new run-time parameters |
999 |
|
(unfortunately) that default to the corresponding exf values; this |
1000 |
|
changes results in most verification experiments so that for backward |
1001 |
|
compatibility these parameter have to be set explicitly to inconsistent |
1002 |
|
values (ugly); |
1003 |
|
- the exf-flag EXF_LWDOWN_WITH_EMISSIVITY then gets rid off the hard-wired |
1004 |
|
emissivities of 0.97 associated with lwdownloc in S/R seaice_solve4temp |
1005 |
|
until we agree how to handle this bug properly |
1006 |
|
o pkg/exf (interpolation): |
1007 |
|
- add a set of run-time param ({inputfield}_interpMethod), one for |
1008 |
|
each interpolated input field, to select the interpolation method |
1009 |
|
with a value of zero switching off the interpolation; |
1010 |
|
- print interpolation parameters to standard output; |
1011 |
|
- refine exf_check.F accordingly ; stop if USE_NO_INTERP_RUNOFF is defined |
1012 |
|
(to use instead runoff_interpMethod=0 in data.exf). |
1013 |
|
- simplify default setting of interpolation parameters (for all fields, |
1014 |
|
set position to match model grid cell-center position, leading to trivial |
1015 |
|
interpolation in case of regular Lat-Lon grid) and fix for the case of |
1016 |
|
non-uniform delY. |
1017 |
|
o pkg/exf/exf_radiation.F: add the factor ocean_emissivity to lwdown, for |
1018 |
|
within #ifdef EXF_LWDOWN_WITH_EMISSIVITY/#endif. Unless ocean_emissivity=1. |
1019 |
|
this factor is absolutely required, because the lw exitance (= out-going |
1020 |
|
long wave radiation) is emissivity*stefanBoltzmann*T^4 + rho*lwdown, |
1021 |
|
where the reflectivity rho = 1-emissivity for conservation reasons. Once |
1022 |
|
this is established, we can remove the CPP-Flag (if we want to). |
1023 |
|
o pkg/seaice: |
1024 |
|
- qualify a stop statement in S/R seaice_check: EXF_READ_EVAP and |
1025 |
|
ALLOW_BULKFORMULAE are only required if SEAICE_EXTERNAL_FLUXES is defined |
1026 |
|
- sort out emissivity and long wave radiation, step 1: make SEAICE_emissivity |
1027 |
|
what the name implies (and not emissivity*BoltzmannConstant), this means |
1028 |
|
the old default value of 5.5e-8 now means |
1029 |
|
5.5e-8/5.670e-8 = 0.97001763668430343479 as ocean_emissivity in |
1030 |
|
S/R exf_readparms |
1031 |
|
- FV-discretization of the B-grid LSOR solver, turn on with |
1032 |
|
SEAICE_LSRBNEW. This new solver also involves a slight re-organization |
1033 |
|
of the calling routine dynsolver, so that B and C-grid code look a |
1034 |
|
little more alike. |
1035 |
|
- fix index bugs in ostres.F (B-grid code) |
1036 |
|
o pkg/seaice/seaice_ocean_budget.F: change (unused) code to make it look better |
1037 |
|
o pkg/seaice/seaice_readparms.F: make ocean_emissivity*stefanBoltzmann the |
1038 |
|
default for SEAICE_emissivity, if useEXF; requires setting a wrong |
1039 |
|
parameter value in global_ocean.cs32x15/input.seaice/data.seaice for |
1040 |
|
backward compatiblity |
1041 |
|
o model/src: use sigmaR in calc_ivdc |
1042 |
|
|
1043 |
|
checkpoint63g (2011/12/16) |
1044 |
|
o pkg/seaice/seaice_growth.F: |
1045 |
|
fix heat conservation of the coupled ocean-seaice system. |
1046 |
|
when SEAICE_HEAT_CONSERV_FIX is defined, the mismatch of |
1047 |
|
solid vs liquid water treatment is resolved via QNET. |
1048 |
|
o pkg/seaice/seaice_growth.F: |
1049 |
|
add missing term for sublimation in saltFlux, which was breaking conservation. |
1050 |
|
o pkg/seaice/seaice_model.F: |
1051 |
|
Remove this patholigical case treatment in evolution branch, |
1052 |
|
where it is un-necessary and breaks conservation. |
1053 |
|
Update global_ocean.cs32x15/results accordingly for |
1054 |
|
output.seaice.txt output_adm.seaice.txt and output_adm.seaice_dynmix.txt |
1055 |
|
o model/src: |
1056 |
|
useRealFreshWaterFlux: remove masking of EmPmP (using maskInC) in |
1057 |
|
external_forcing_surf.F and apply mask directly in S/R SOLVE_FOR_PRESSURE. |
1058 |
|
o pkg/obcs: |
1059 |
|
modify Stevens boundary conditions so that S/R obcs_calc_stevens computes |
1060 |
|
tendency for temperature and salinity that is then applied to update |
1061 |
|
theta and salt in S/R obcs_apply_ts; this requires proper masking of all |
1062 |
|
horizontal advection and diffusion contributions on the boundary (in |
1063 |
|
S/Rs gad_advection, gad_calc_rhs, gmredi_rtransport, gmredi_calc_diff), and |
1064 |
|
removing the mask in S/R timestep_tracer. The advantage of this approach is |
1065 |
|
that the updated tracer values include contributions from the surface |
1066 |
|
forcing and vertical diffusion. |
1067 |
|
This modification changes exp4.stevens because of vertical diffusion. |
1068 |
|
o model/src: |
1069 |
|
change non-curvilinear grid setting to also work with exch2 + blank tiles: |
1070 |
|
- move delX,delY to new header file (SET_GRID.h) and adjust length to 1rst |
1071 |
|
face dimensions. |
1072 |
|
- new S/R INI_LOCAL_GRID to define model tile-local grid (using same units |
1073 |
|
as delX,delY) and called from S/R INI_CARTESIAN_GRID, INI_CYLINDER_GRID |
1074 |
|
and INI_SPHERICAL_POLAR_GRID. |
1075 |
|
change test-experiment global_ocean.90x40x15: |
1076 |
|
- split the domain in many more tiles (9x4 tiles of 10x10 each); |
1077 |
|
- switch to exch2 and remove 1 blank-tile (but only for MPI test). |
1078 |
|
- update results of both test (std & dwnslp). |
1079 |
|
o pkg/diagnostics/diagstat_calc.F: fix problem with non-initialised halos in |
1080 |
|
tmp-fields for TARGET_NEC_SX at the cost of extra loops and less vectorisation |
1081 |
|
o model/src: |
1082 |
|
- take sigma-coords related code out of update_etah.F into new S/R UPDATE_ETAWS |
1083 |
|
and fix initialisation when using sigma-coords. |
1084 |
|
o model/src: |
1085 |
|
- improve Gael's changes from Jan 14, 2011 "addition of an 'update' of hfacc etc. |
1086 |
|
to their current value at the beginning of forward_step.F": add a new run-time |
1087 |
|
switch "doResetHFactors" (default=false) and use it to do (or not) the resetting. |
1088 |
|
- set doResetHFactors=T in the following test exp: |
1089 |
|
global_ocean.90x40x15/input (rStar) & input.dwnslp (linFS), |
1090 |
|
global_ocean.cs32x15/input.seaice (rStar), hs94.cs-32x32x5/input.impIGW (linFS) |
1091 |
|
and tutorial_global_oce_in_p (nonLinFS in p) |
1092 |
|
o pkg/ecco |
1093 |
|
- added IES cost term |
1094 |
|
o verification/advect_xz |
1095 |
|
- scale down the depth by 1/100 (from 200 km to just 2.km) |
1096 |
|
- add secondary test using divergent flow, r-star and implicit vertical advection. |
1097 |
|
- add note (check_conserve_TS.txt) and matlab script to check global |
1098 |
|
conservation of T & S with NonLin-FreeSurf and Adams-Bashforth. |
1099 |
|
o pkg/rbcs: |
1100 |
|
- add missing relaxUFile & relaxVFile in namelist (in rbcs_readparms.F). |
1101 |
|
o pkg/obcs (Stevens BCs): |
1102 |
|
- fix various problems in the phase speed computations of the Stevens |
1103 |
|
boundary conditions: |
1104 |
|
- sign error for western and southern boundaries, |
1105 |
|
- time tendencies were underestimated by factor of dTracerLev and some just |
1106 |
|
zero (for multidim advection schemes) -> now estimate time tendencies |
1107 |
|
from difference between previous (n) time level and the one before (n-1); |
1108 |
|
this requires storing a field per tracer per boundary, also in the pickup |
1109 |
|
files. |
1110 |
|
- replace deltaTmom in cflMer/Zon computation with dTracerLev |
1111 |
|
- set default phase velocity to cflMer/Zon |
1112 |
|
- update verification experiment exp4.stevens |
1113 |
|
o pkg/generic_advdiff (implicit_r): |
1114 |
|
- fix implicit vertical advection conservation with AB and/or NonLin-FreeSurf |
1115 |
|
(remove Tr*d/dz(w) in gad_implicit_r.F and add it in gad_calc_rhs.F); |
1116 |
|
- fix implicit vertical diffusion or advection with NonLin FreeSurf |
1117 |
|
(was using current recip_hFacC instead of future recip_hFacC): set future |
1118 |
|
recip_hFac in thermodynamics.F and pass & use it in all vertical implicit |
1119 |
|
solver S/R ; |
1120 |
|
- fill diagnostics of vertical advective fluxes (computed from tendency) |
1121 |
|
when using implicit vert. advection; |
1122 |
|
- update output of exp that uses NonLin-FreeSurf & implicit vertical diffusion: |
1123 |
|
global_ocean.90x40x15 (fwd+AD), global_ocean.cs32x15 (4 fwd +4 AD), |
1124 |
|
tutorial_global_oce_in_p (fwd), tutorial_tracer_adjsens (AD) |
1125 |
|
and cpl_aim+ocn (used in ocean); to be fixed: bottom_ctrl_5x5. |
1126 |
|
o pkg/diagnostics: |
1127 |
|
- add diags for T & S tendency which goes through Adams-Bashforth. |
1128 |
|
|
1129 |
|
checkpoint63f (2011/11/30) |
1130 |
|
o pkg/obcs/obcs_calc_stevens.F: |
1131 |
|
- join some loops to reduce number of if-statements |
1132 |
|
- fix a bug in code for northern and southern boundaries. |
1133 |
|
o pkg/seaice/seaice_init_varia.F |
1134 |
|
- undo resetting UVM=0 that has slipped in unnoticed in r1.49 and |
1135 |
|
caused zero ice-velocities for the B-grid code |
1136 |
|
o pkg/seaice/seaice_init_fixed.F |
1137 |
|
- fix a bug in the initialisation of the second metric terms |
1138 |
|
k2AtC and k2AtZ: k2AtC was set to the value of k2AtZ and k2AtZ was 0. |
1139 |
|
- this affects all seaice configurations with useCurviLinearGrid = .true. |
1140 |
|
and SEAICEuseDynamics = .true. (default) and SEAICEuseMetricTerms |
1141 |
|
= .true. (default) |
1142 |
|
- the only verification experiment affected is global_ocean.cs32x15: |
1143 |
|
icedyn, seaice, and seaice_dynmix (AD) |
1144 |
|
o pkg/thsice/thsice_diagnostics_state.F |
1145 |
|
- only for TARGET_NEC_SX: initialise all of tmpFld, so that overlap does |
1146 |
|
not contain garbage that can lead to surprises in diagstats_calc.F |
1147 |
|
o model/src: |
1148 |
|
- move bi,bj loops inside INTEGR_CONTINUITY ; |
1149 |
|
- call INTEGR_CONTINUITY directly from forward_step.F |
1150 |
|
(previously called from momentum_correction_step.F) |
1151 |
|
- call UPDATE_ETAH from integr_continuity.F and update initialise_varia.F, |
1152 |
|
forward_step.F and pkg/ecco/the_main_loop.F |
1153 |
|
o model/src/do_oceanic_phys.F: |
1154 |
|
- reset addMass to zero only if useICEFRONT ; |
1155 |
|
- separated k-loop for rhoInSitu calculation ; |
1156 |
|
- add 2 diagnostics for wVel*delta^k(Rho) at constant pressure (= WdRHO_P) |
1157 |
|
and at constant T,S (= WdRHOdP); |
1158 |
|
NOTE: change diagnostics WRHOMASS : this is now simply the Vertical transport |
1159 |
|
of In-Situ Density (Anomaly) (=wVel*rho^bar-k) (previously was Vertical |
1160 |
|
transport of locally-referenced potential density); |
1161 |
|
to recover old diag: (WRHOMASS)_old = (WRHOMASS)_new -1/2 * (WdRHOdP) |
1162 |
|
o gluemnc: |
1163 |
|
- replace ncrcat with ncks when adding variables to *.glob.nc |
1164 |
|
- change the way variables without record dimensions are handled so that |
1165 |
|
files that have variables with and without record variables can be glued: |
1166 |
|
add a record dimension with a new unique name to be able to differentiate |
1167 |
|
between variable with and without a-prior record dimensions. Remove |
1168 |
|
this record dimension before adding the variable to the *.glob.nc file. |
1169 |
|
This procedure automatically works also for grid.*.nc files that do not |
1170 |
|
have any record dimension. |
1171 |
|
o verification/cpl_aim+ocn: |
1172 |
|
- generate new reference output on baudelaire (gfortran, mpich2) (previously |
1173 |
|
from old aces-grid cluster, ao + ifort) and update 'run_cpl_test' script. |
1174 |
|
o pkg/obcs/obcs_calc_stevens.F: |
1175 |
|
- add code for northern and southern boundaries. |
1176 |
|
- reorder some loops to save some operations and so that drFBar can be scalar |
1177 |
|
o model/src: |
1178 |
|
- re-arrange extrapolation calculation in Adams-Bashforth S/R (affects machine |
1179 |
|
truncation) and add diagnostics for Adams-Bashforth tendency increment. |
1180 |
|
- update test-exp. AD output: |
1181 |
|
1D_ocean_ice_column, bottom_ctrl_5x5, lab_sea.noseaicedyn, obcs_ctrl; |
1182 |
|
and FWD output: exp4 (x3), fizhi-cs-32x32x40, fizhi-cs-aqualev20, |
1183 |
|
front_relax (x3), global_ocean.90x40x15 (x2), global_ocean.cs32x15 (x5), |
1184 |
|
global_ocean_ebm, ideal_2D_oce, internal_wave, isomip (x3), lab_sea (x4), |
1185 |
|
matrix_example, MLAdjust (x5), natl_box (x2), tutorial_advection_in_gyre, |
1186 |
|
tutorial_global_oce_in_p, tutorial_global_oce_latlon and |
1187 |
|
tutorial_plume_on_slope. |
1188 |
|
o pkg/diagnostics/diagstats_calc.F: add special code for TARGET_NEC_SX |
1189 |
|
that vectorises. This code uses 3 extra 2D fields, F90 array assignments |
1190 |
|
and F90 intrinsic routines SUM, MINVAL, MAXVAL, but since TARGET_NEC_SX |
1191 |
|
does not even come with an F77 compiler anymore, this should not be |
1192 |
|
a limitation |
1193 |
|
|
1194 |
|
checkpoint63e (2011/11/06) |
1195 |
|
o pkg/seaice adjoint of LSR: |
1196 |
|
- substantially rework STORE dirs. |
1197 |
|
- add new flag SEAICE_LSR_ADJOINT_ITER and parameter SOLV_MAX_FIXED |
1198 |
|
this flag will do SOLV_MAX_ITERS but only store SOLV_MAX_FIXED fields |
1199 |
|
and only perform SOLV_MAX_FIXED adjoint iters. |
1200 |
|
If flag SEAICE_LSR_ADJOINT_ITER is undef-ined, the store in the iteration |
1201 |
|
always goes to same record, meaning that the last forward field will be |
1202 |
|
the one used in the adjoint - an approximation. |
1203 |
|
- now produces non-zero LSR-related gradients, and acceptable gradients for |
1204 |
|
An's zonal (x-z) test case (to be checked in) |
1205 |
|
- lab_sea: produces acceptable gradients for lab_sea with forward/backward LSR |
1206 |
|
(but these are screwed up by the thermodynamics in the first place); |
1207 |
|
reduce from 4 to 3 timesteps to avoid t.d. issue as temporary solution; |
1208 |
|
output_adm.txt also reflects changes to seaice_lsr |
1209 |
|
- global_ocean.cs32x15: |
1210 |
|
* the "default" input_ad.seaice has been re-defined to always exclude dynamics |
1211 |
|
(output_adm.seaice.txt was updated) |
1212 |
|
* new input_ad.seaice_dynmix tests a forward LSR / backward free-drift |
1213 |
|
(output_adm.seaice_dynmix.txt) |
1214 |
|
o test-exp. exp4+stevens: |
1215 |
|
- switch off pTracers and nonHydrostatic (restart problems with NH); update output |
1216 |
|
o pkg/seaice (lsr solver): |
1217 |
|
- put coriolis term into FORCEX,Y (as in seaice_evp.F; previously in rhsU,V only); |
1218 |
|
update output (change truncation) of FWD test exp lab_sea & global_ocean.cs32x15 |
1219 |
|
- fix SEAICEuseFREEDRIFT for new OBCS version (ndef OBCS_UVICE_OLD) |
1220 |
|
- new parameter "LSR_mixIniGuess" allows (if=2,4) to mix free-drift solution |
1221 |
|
into initial guess for linearised solver; use weight ~ (1/err)^2,4. |
1222 |
|
- compute (with new S/R) and print initial (+ free-driff sol. if available) |
1223 |
|
and final residual of linearised system. |
1224 |
|
o pkg/pp81: add diagnostics for vertical eddy viscosity and diffusivity |
1225 |
|
(requires new subroutines pp81_init_fixed.F/pp81_diagnostics_init.F) |
1226 |
|
o pkg/obcs: Stevens BCs (still only for east and west) |
1227 |
|
- fix Stevens BCs by using intermediate velocities of previous |
1228 |
|
time step as "baroclinic" contribution rather velocity of one grid cell |
1229 |
|
inside the boundaries; more consistent but makes the entire treatment |
1230 |
|
more complicated, requires new pickup files and two new S/Rs: |
1231 |
|
obcs_copy_uv_n and obcs_save_uv_n |
1232 |
|
keep old version with flag OBCS_STEVENS_USE_INTERIOR_VELOCITY |
1233 |
|
(undefined by default) |
1234 |
|
- silently move computation of some grid factors into separate loops to |
1235 |
|
save some recomputations |
1236 |
|
- add a hack for non-linear free surface |
1237 |
|
- add a verification experiment to exp4 that tests this code |
1238 |
|
o make gluemnc/xplodemnc work with (moron) non-GNU shell tools |
1239 |
|
|
1240 |
|
checkpoint63d (2011/10/25) |
1241 |
|
o pkg/obcs + seaice-dynamics: |
1242 |
|
- new implementation calls OBCS_APPLY_UVICE (in seaice_dynsolver.F) before |
1243 |
|
solving for uIce,vIce and prevent SEAICE_LSR or SEAICE_EVP to modify OB |
1244 |
|
values (consistent with an implicit method). |
1245 |
|
- add option OBCS_UVICE_OLD (in OBCS_OPTIONS.h, defined by default) to |
1246 |
|
return to older implementation (same results as before); note that none of |
1247 |
|
the "experimental" OBCS_SEAICE options are implemented in the new version. |
1248 |
|
- switch test exp. seaice_obcs to use new version (update output). |
1249 |
|
o pkg/seaice: |
1250 |
|
- add few stop (B-grid & CS-exch, B-grid & obcs, AUTODIFF & SEAICE_maskRHS) |
1251 |
|
- add parenthesis in Water-Ice drag calculations (evp & lsr); |
1252 |
|
- in seaice_lsr.F, simplify and fix AA3 setting in case sNx or sNy=1; |
1253 |
|
and fix case SEAICE_waterDrag=0 (but unstable if no or little seaice) |
1254 |
|
- update output of test exp lab_sea that uses evp or lsr (fwd+ad) |
1255 |
|
and global_ocean.cs32x15 (icedyn & seaice fwd+ad). |
1256 |
|
o pkg/layers: |
1257 |
|
- Include potential density as new coordinate (Thanks to David Munday) |
1258 |
|
o pkg/dic: |
1259 |
|
- Remove subroutine CALC_PCO2_APPROX_CO3 from carbon_chem.F and add |
1260 |
|
carbonate computation/output to CALC_PCO2_APPROX |
1261 |
|
o pkg/generic_advdiff |
1262 |
|
- make TAF generate code that vectorizes: |
1263 |
|
gad_dst3fl_adv_*.F, gad_u3c4_impl_r.F |
1264 |
|
o model/src: |
1265 |
|
- improve vectorization also for adjoint code of |
1266 |
|
solve_tridiagonal and solve_pentadiagonal. As this requires up to |
1267 |
|
5 new 3D-fields, this code is only turned on for the combination of |
1268 |
|
ALLOW_AUTODIFF_TAMC and TARGET_NEC_SX. Also remove error message |
1269 |
|
for singular matrix check (only for TARGET_NEC_SX). |
1270 |
|
- remove hard-wiring of CPP-flag ALLOW_SOLVERS_KLOOPINSIDE to |
1271 |
|
ALLOW_AUTODIFF. This flag now needs to be set outside of |
1272 |
|
solve_tri/pentadiagonal.F |
1273 |
|
o pkg/exf: |
1274 |
|
- fix previous modif in exf_readparms.F (revision 1.66: treatment of |
1275 |
|
optional namelist EXF_NML_OBCS). |
1276 |
|
- add many debug-print (to trace down where pkg/cal error msg comes from) |
1277 |
o pkg/mom_vecinv/mom_vi_hdissip.F |
o pkg/mom_vecinv/mom_vi_hdissip.F |
1278 |
- rearrange loops so that vectorization is improved and in particular |
- rearrange loops so that vectorization is improved and in particular |
1279 |
the adjoint routine vectorizes. The code is a little longer now, but I |
the adjoint routine vectorizes. The code is a little longer now, but I |
1280 |
think easier to read. |
think easier to read. |
1281 |
Note: CPP-flag MOM_VI_ORIGINAL_VISCA4 is not tested. |
Note: CPP-flag MOM_VI_ORIGINAL_VISCA4 is not tested. |
1282 |
o pkg/seaice: |
o pkg/seaice: |
1283 |
- allow to specify initial seaice velocity from files (uIceFile,vIceFile) |
- allow to specify initial seaice velocity from files (uIceFile,vIceFile) |