4 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
5 |
============================== |
============================== |
6 |
|
|
7 |
|
o pkg/thsice: |
8 |
|
- re-anable balanceEmPmR when useSEAICE=T with useThSIce=T. |
9 |
|
- option to correct ocean fresh-water flux for imbalance of Atmos+Land |
10 |
|
fresh-water flux: uniform correction or proportional to local precip. |
11 |
|
o pkg/ctrl: |
12 |
|
- new header file CTRL_SIZE.h for parameters maxCtrlArr2D, maxCtrlArr3D |
13 |
|
- redefine CPP option ALLOW_BOTTOMDRAG_CONTROL to |
14 |
|
ALLOW_BOTTOMDRAG_CONTROL_NONGENERIC |
15 |
|
- verification/hs94.1x64x5/ tests ALLOW_GENARR{2D,3D}_CONTROL |
16 |
|
|
17 |
|
checkpoint63q (2012/07/30) |
18 |
|
o pkg/autodiff: |
19 |
|
note about monitor and snap-shot of ad-variables: |
20 |
|
a) need to apply ADEXCH before monitor or output of ad-vars to include |
21 |
|
overlap contributions (was done for adjDump but missing for monitor). |
22 |
|
b) since ADEXCH implies adding overlap values to interior array, machine |
23 |
|
truncation error is expected when adding an ADEXCH call in the middle of |
24 |
|
the time-stepping (although mathematically identical). |
25 |
|
modifs: |
26 |
|
- fix some incorrect type (RL/RS, UV/scalar) in ADEXCH calls and ad-vars |
27 |
|
declaration (adEddyPsiX,adEddyPsiY in adcommon.h). |
28 |
|
- add new run-time params (dumpAdVarExch & mon_AdVarExch) to select how to |
29 |
|
apply ADEXCH before monitor or snap-shot of ad-variables: |
30 |
|
=0 : no ADEXCH ; =1 : do ADEXCH ; =2 : apply ADEXCH on a local copy. |
31 |
|
- implement new method (copy+adexch) for snap-shot and monitor of main model |
32 |
|
ad-vars (initially not used by default -> output unchanged). |
33 |
|
- change default (set to 2) to use new method and update few output_adm.txt |
34 |
|
with accurate ad-monitor (global_ocean.90x40x15, tutorial_global_oce_biogeo |
35 |
|
and both tutorial_tracer_adjsens tests). |
36 |
|
- switch 2 global_ocean.cs32x15 AD-tests to AB-2 (seaice_dynmix & thsice) |
37 |
|
and turn off momDissip_In_AB in the 2 other (needed for stable AB-3). |
38 |
|
o pkg/ctrl: |
39 |
|
enable the use of a variable number (maxCtrlArr2D, maxCtrlArr3D) of |
40 |
|
generic control variables via |
41 |
|
#define ALLOW_GENARR2D_CONTROL and/or #define ALLOW_GENARR3D_CONTROL |
42 |
|
Only required customization should be S/R ctrl_map_ini_genarr.F |
43 |
|
(routine in repository contains example) |
44 |
|
o pkg/exf: |
45 |
|
- fix and refine checking of surface-relaxation setting when using EXF |
46 |
|
o verification (adjoint exp): |
47 |
|
- reduce length of simulation by half for experiment global_ocean.cs32x15 |
48 |
|
(std, seaice & seaice_dynmix) and exp tutorial_dic_adjoffline; update output. |
49 |
|
o pkg/mom_common (mom_u/v_bottomdrag), pkg/shelfice (shelfice_u/v_drag) |
50 |
|
- replace if statements KE.ne.0 with KE.gt.0 to avoid sqrt of negative numbers |
51 |
|
of KE that can occur with aggressive optimization |
52 |
|
+ put quadratic bottom drag calculation in separated loops (only done if |
53 |
|
using bottomDragQuadratic) |
54 |
|
o model/src: CG2D_NSA |
55 |
|
- clean-up: stop if cg2dNormaliseRHS=T and AUTODIFF |
56 |
|
and replace GLOBAL_SUM calls with GLOBAL_SUM_TILE. |
57 |
|
- in test-exp that uses CG2D_NSA (bottom_ctrl & tutorial_tracer_adjsens), |
58 |
|
change cg2d residual target to be expressed in Wunits (=> do not normalise |
59 |
|
RHS ; this fix global_max Pb with TAF); update output. |
60 |
|
o verification/tutorial_tracer_adjsens: |
61 |
|
improve parameter consistency (and get rid of warnings): |
62 |
|
- use deltaTFreeSurf=deltaTtracer (better with NonLin FreeSurf); |
63 |
|
- switch Temp Adv Scheme from 81 to 80 (no limiter since theta < 0 is physical) |
64 |
|
- use staggerTimeStep (for stability) in som81; use default rotation period. |
65 |
|
o pkg/exch2: |
66 |
|
- with empty facet: fix compact-format (W2_mapIO=1) definition (was previously |
67 |
|
reset to 0 = 1 long line in X) |
68 |
|
o pkg/mdsio: |
69 |
|
- fix reading of dimensions from meta file for large-size domain |
70 |
|
o model/src: |
71 |
|
- fix saltIsActiveTr setting for IdealGas eosType (eosType is 6 character long); |
72 |
|
- to refine CS-grid check, add logical flag "hasWetCSCorners": true if using |
73 |
|
Cubed-Sphere Exch with CS-corners inside the domain; |
74 |
|
make new S/R INI_GLOBAL_DOMAIN from code in ini_linear_phisurf.F to calculate |
75 |
|
globalArea and to set hasWetCSCorners. |
76 |
|
- add warning if usingCurvilinearGrid and momAdvection in flux-form (missing |
77 |
|
metric terms); stop if hasWetCSCorners=T and momAdvection in flux-form. |
78 |
|
- re-activate the stop if useCDscheme and hasWetCSCorners=T (previously if |
79 |
|
useCDscheme and useCubedSphereExchange). |
80 |
|
o pkg/seaice/seaice_growth.F : |
81 |
|
- compute the SEAICE_DISABLE_HEATCONSFIX term accordingly |
82 |
|
for the (.NOT.useRealFreshWaterFlux).OR.(nonlinFreeSurf.EQ.0) |
83 |
|
when the previous code was not correct. |
84 |
|
- in all cases diagnoze the boundary condition mismatch to SIaaflux |
85 |
|
- in the case of useRealFreshWaterFlux.AND.(nonlinFreeSurf.GT.0) add |
86 |
|
that term to QNET (essentially to set the melt/freeze water to 0degC). |
87 |
|
when SEAICEheatConsFix is true, so that the ocean+ice budget heat is closed. |
88 |
|
o verification/global_ocean.cs32x15/input.seaice : |
89 |
|
- add SEAICEheatConsFix in data.seaice, ensuring closed ocean+ice heat budget. |
90 |
|
- add diagnostics (commented out) for budget test using my matlab codes (gcmfaces). |
91 |
|
- update results (output.seaice.txt) accordingly. |
92 |
|
o verification/lab_sea/input/data.seaice : |
93 |
|
- remove SEAICEheatConsFix, which is now be tested |
94 |
|
in global_ocean.cs32x15. The presently coded fix |
95 |
|
is only consistent with real fresh water + nlfs. |
96 |
|
- update results (output.txt) accordingly. |
97 |
|
o pkg/salt_plume: added diagnotic for salt tendency (oceSPtnd). |
98 |
|
o AD_CONFIG.h file: |
99 |
|
- in genmake2: remove "#undef ALLOW_ECCO_OPTIMIZATION" from AD_CONFIG.h; |
100 |
|
- remove #include "AD_CONFIG.h" from all included OPTIONS files and |
101 |
|
add it explicitly in each fortran src file where it is needed. |
102 |
|
motivation: header file AD_CONFIG.h is generated/modified after "make depend" |
103 |
|
thus introducing potentially wrong dependency; therefore, better to include |
104 |
|
it only where it's really needed, and after any other included OPTIONS file |
105 |
|
(less tempting to bring conditional includes according to ALLOW_ADJOINT_RUN |
106 |
|
or ALLOW_TANGENTLINEAR_RUN). |
107 |
|
|
108 |
|
checkpoint63p (2012/07/06) |
109 |
|
o pkg/autodiff: |
110 |
|
- rename S/R INI_AUTODIFF to AUTODIFF_INIT_VARIA ; |
111 |
|
- create new header file "AUTODIFF_PARAMS.h" for pkg/autodiff parameters |
112 |
|
which are now read from new parameter file "data.autodiff" (inAdExact |
113 |
|
previously in "data", use{KPP/GMRedi/SEAICE}inAdMode previously in data.pkg) |
114 |
|
- add file "data.autodiff" in all verification/*/input_ad/ directories. |
115 |
|
- add run-time parameter to turn On/Off SEAICEuseFREEDRIFT in adjoint/forward |
116 |
|
mode (used in global_ocean.cs32x15.seaice_dynmix test exp.) |
117 |
|
o verification/tutorial_tracer_adjsens/ |
118 |
|
- add extra sub-verification which uses "SOM" advection scheme |
119 |
|
(2nd Order-Moment Advection Scheme, Prather Limiter) with |
120 |
|
#define GAD_ALLOW_TS_SOM_ADV in GAD_OPTIONS.h |
121 |
|
and in data: |
122 |
|
multiDimAdvection=.TRUE., |
123 |
|
tempAdvScheme=81, |
124 |
|
saltAdvScheme=81, |
125 |
|
o model/src: |
126 |
|
- introduce ALLOW_BALANCE_RELAX which allow the removal |
127 |
|
of the global mean of relaxation terms by setting |
128 |
|
balanceThetaClimRelax and balanceSaltClimRelax |
129 |
|
- disable balanceEmPmR and balanceQnet in the case when useSeaice. |
130 |
|
This case is now treated appropriately in seaice_growth.F |
131 |
|
o pkg/seaice: |
132 |
|
- do balanceEmPmR and balanceQnet based on the net fluxes |
133 |
|
(SItflux) that enter the ocean+ice system rather than QNET |
134 |
|
and EMPMR. Works best in real fresh water. |
135 |
|
- SIatmFW, SIatmQnet, and SItflux computations had to be moved |
136 |
|
out of the diagnotics bloc, and stored in global arrays. |
137 |
|
|
138 |
|
checkpoint63o (2012/06/29) |
139 |
|
o vertical mixing schemes: |
140 |
|
- remove unused S/R CALC_BUOYANCY (left from branch-atmos merging) |
141 |
|
- replace computation of density (in Function STATE1, opps_calc.F) by a call |
142 |
|
to S/R FIND_RHO_SCALAR |
143 |
|
- pass sigmaR to S/R PP81_CALC, MY82_CALC (but not yet used) and to GGL90_CALC |
144 |
|
and use it for N^2 calculation in ggl90_calc.F (save 2 density computations); |
145 |
|
affects output (vermix.ggl90) at truncation level. |
146 |
|
o testreport: |
147 |
|
- change adjustment of SIZE.h for mpi+mth testing: preserve enough tiles |
148 |
|
(nSx,nSy) for the number of threads (from eedata.mth) |
149 |
|
o model/src: |
150 |
|
- remove 2nd set of calls at the top of initialise_varia.F (Adjoint part) |
151 |
|
related to Non_Lin Free-Surf or Bottom-Control (but not longer necessary). |
152 |
|
o verification/tutorial_tracer_adjsens/ |
153 |
|
- compile with ALLOW_CG2D_NSA |
154 |
|
- run with nonlinFreeSurf=3 |
155 |
|
- update reference output_adm.txt |
156 |
|
o tools/genmake2: |
157 |
|
- add "-DHAVE_NETCDF" to the DEFINES list when NetCDF is available ; |
158 |
|
refine usePROFILES switching off (only when NetCDF is not available). |
159 |
|
o verification (atmospheric set-up): |
160 |
|
- change eosType='IDEALGAS' to eosType='IDEALG' to match variable declaration |
161 |
|
o pkg/profiles : refine profiles and files handling. |
162 |
|
o model/src and pkg/ecco : introduce run-time switch usePROFILES. |
163 |
|
o model/src/the_main_loop.F, forward_step.F and pkg/exf/EXF_OPTIONS.h : |
164 |
|
- changes to use model/src/the_main_loop.F rather than the ecco version |
165 |
|
in ecco type adjoint runs. |
166 |
|
o model/src: |
167 |
|
- move computation of recip of grid distance & area from INI_MASKS_ETC to INI_GRID |
168 |
|
- move setting of h0Fac[C,W,S] from INI_LINEAR_PHISURF to INI_MASKS_ETC |
169 |
|
- update_masks_etc.F: rename SMOOTH*_R4,R8 function to the corresponding type (RS,RL) |
170 |
|
o model/src: |
171 |
|
- fix S/R UPDATE_CG2D for ALLOW_DEPTH_CONTROL case (was empty before); |
172 |
|
update output of test experiment bottom_ctrl_5x5. |
173 |
|
o pkg/generic_advdiff: |
174 |
|
- implement vertical bi-harmonic diffusion (for now, only for T & S and |
175 |
|
only explicitly). |
176 |
|
o verification/bottom_ctrl_5x5/ |
177 |
|
update reference output (results changed quite some time ago, |
178 |
|
not sure when; but new reference results look ok in terms of grdchk) |
179 |
|
o verification/1D_ocean_ice_column/ |
180 |
|
For verification purposes, the exact adjoint rather than an approx. |
181 |
|
should be used (e.g., to comare against F.D. or TLM). |
182 |
|
removed autodiff_inadmode*_ad.F routines in code_ad/ |
183 |
|
Update results/ accordingly. |
184 |
|
o pkg/cal: |
185 |
|
- change cal_getdate.F so that output date is function of input time only |
186 |
|
(less ambiguous, not mixed up with input iteration number). |
187 |
|
- fix calendarDumps for non-zero phase-lag (S/R CAL_TIME2DUMP). |
188 |
|
o pkg/diagnostics: |
189 |
|
- implement calendarDumps also for stats-diags |
190 |
|
- implement calendarDumps also for snap-shot output (negative freq) |
191 |
|
- for now, ignores calendarDumps if non-zero phase-lag. |
192 |
|
|
193 |
|
checkpoint63n (2012/06/04) |
194 |
|
o model/src: |
195 |
|
- recompute rSurfW & S to ensure consistent column thickness (needed for rStar) |
196 |
|
between Sum_k(hFac*drF) and rSurf-rLow, for now only when using pkg/shelfice; |
197 |
|
- use h0FacC, W & S to compute hFac_surfC, W & S in calc_surf_dr.F (more |
198 |
|
similar to calc_r_star.F); update results of experiment internal_wave and |
199 |
|
tutorial_global_oce_in_p (affected @ truncation level). |
200 |
|
o pkg/shelfice: |
201 |
|
- add parameterisation of advective-diffusive flux into the ice shelf |
202 |
|
according to Holland and Jenkins (1999), eq.22-33, as suggested by Jan De Ryd |
203 |
|
- simply eps3/3a computation |
204 |
|
o eesupp: |
205 |
|
- add new CPP option (DISCONNECTED_TILES) to disconnect tiles (no exchange |
206 |
|
between tiles, just fill-in edges assuming locally periodic subdomain) and |
207 |
|
add corresponding exchange subroutines (EXCH0); |
208 |
|
- new CG-solver version (_EX0) for disconnected-tiles special case. |
209 |
|
o model/src (CG-solvers): |
210 |
|
- modify Single-Reduction version (CG2D_SR) in case cg2dMaxIters is reached |
211 |
|
(similar final residual output and max-Iter number as with standard version); |
212 |
|
- allows to store and use lowest-residual solution (standard & _SR versions); |
213 |
|
- cleaning up (remove old, commented out code, LETS_MAKE_JAM ...). |
214 |
|
- by default, also compile CG2D_SR version (define ALLOW_SRCG in CPP_OPTIONS.h) |
215 |
|
o eesup/src: |
216 |
|
- global_vec_sum.F cleaning: add a BARRIER ; update comments ; remove ndef |
217 |
|
ALWAYS_USE_MPI (no longer needed after fixing usingMPI settings - March 2012) |
218 |
|
o pkg/salt_plume added to lab_sea/input_ad.noseaicedyn |
219 |
|
|
220 |
|
checkpoint63m (2012/04/27) |
221 |
|
o pkg/shelfice: |
222 |
|
- account for partical cells in surface level below ice-shelf for hydrostatic |
223 |
|
pressure calculion (set uniformFreeSurfLev to F); update isomip experiments. |
224 |
|
o ad test experiments: |
225 |
|
- update and rename to "ad_optfile.local" local AD-optfile specific to a |
226 |
|
particular adjoint verification experiment (in verification/*/code_ad/) |
227 |
|
o pkg/seaice & pkg/exf: |
228 |
|
- Part two of ECCO-CLIVAR code merges. |
229 |
|
Completes merge of seaice area relax. code. |
230 |
|
Requires unified CPP option EXF_ALLOW_SEAICE_RELAX |
231 |
|
to be set in EXF_OPTIONS.h since this pkg controls its I/O |
232 |
|
(and avoids second CPP option in SEAICE_OPTIONS.h) |
233 |
|
- change default value of costIceStart & costIceEnd to a valid date. |
234 |
|
- add snowPrecip to HSNOW in seaice_growth |
235 |
|
o pkg/ptracers: |
236 |
|
- move ptracers summary and setup check into separated S/R ; add a check for |
237 |
|
valid PTRACERS_Iter0 value (error+stop or just warning). |
238 |
|
o pkg/cal (+ pkg/ecco): |
239 |
|
- avoid calling a S/R with 2 times the same arg (FWD: unless both are only used |
240 |
|
as input; and for AD, unless both are not differentiable (e.g., k index)). |
241 |
|
o model/src: |
242 |
|
- define internal flag "uniformFreeSurfLev" (=T when free-surface level-index |
243 |
|
is everywhere k=1) to speed-up some calculations (by pass kSurf reference); |
244 |
|
temporary set to usingZCoords value (to be changed to F when using shelfice). |
245 |
|
- hydrostatic pressure: implement partial cell near the surface also for |
246 |
|
Z-coordinates (for the case where uniformFreeSurfLev=F); replace test on |
247 |
|
usingZCoords with test on uniformFreeSurfLev where needed (in NonHyd code). |
248 |
|
o pkg/ctrl/ctrl_init.F: |
249 |
|
- add a formal parameter to s/r ctrl_init_rec to print |
250 |
|
xx_$(ctrl_valiable)_file for easier debugging |
251 |
|
- move call of cal_* routines into the if-block where they are really |
252 |
|
needed so that xx_$(ctrl_variable)_period can be 0 or -12 without having |
253 |
|
to specify a startdate |
254 |
|
o pkg/seaice: |
255 |
|
- update comments in SEAICE_OPTIONS.h (+ similar update of lab_sea/code and |
256 |
|
global_ocean.cs32x15/code versions). |
257 |
|
- turn on seaice Multi-category (in data.seaice, SEAICE_multDim=7) for |
258 |
|
verification experiment global_ocean.cs32x15.seaice and update results. |
259 |
|
o pkg/exf: |
260 |
|
- new S/R EXF_GETFIELD_START to return starting-time of a forcing field |
261 |
|
(according to useExfYearlyFields). |
262 |
|
o pkg/cal: |
263 |
|
- move calendar setting & summary (previously in cal_readparams.F) into new |
264 |
|
S/R CAL_INIT_FIXED to be called later in the initialisation process, once |
265 |
|
length of the simulation is set (after SET_PARMS call); This is an attempt |
266 |
|
to fix Pb when, e.g.: pkg/longstep, modifies the length of the run. |
267 |
|
- For safety: keep a record of pkg/cal parameter settings status |
268 |
|
to allow to STOP if one pkg/cal S/R is called too early. |
269 |
|
- add a stop in all basic pkg/cal subroutines if called before setting is done. |
270 |
|
- modify cal_getdate.F: stop when nothing was returned; |
271 |
|
if called with myIter=-1, simply retrun modelBaseDate (1,2 = startDate_1,_2); |
272 |
|
this allows to remove "cal.h" from seaice_read_parms.F & init_mnc_vars.F. |
273 |
|
|
274 |
|
checkpoint63l (2012/04/04) |
275 |
|
o pkg/bbl: |
276 |
|
- fix call to FIND_RHO_2D in bbl_calc_rho.F (case multi-tiles per proc); |
277 |
|
update global_with_exf output.yearly.txt file. |
278 |
|
o pkg/cal: |
279 |
|
- start to clean-up cal_set.F ; fix for the case where baseTime is non-zero. |
280 |
|
- remove 2 unsused and very short (1 line only) functions ; |
281 |
|
- remove unsused subroutines from AD-Diff list. |
282 |
|
o model/src: |
283 |
|
- start to implement clean wait to stop in initialization S/R (config_check.F, |
284 |
|
packages_check.F, ini_parms.F and pkg/obcs/obcs_check.F + pkg/mypackage) |
285 |
|
o folowing PRINT_MESSAGE changes: |
286 |
|
- remove reminder multiple print to the same file (STDOUT) when multi-threads. |
287 |
|
o pkg/bbl: |
288 |
|
- modification so it can work with z* |
289 |
|
- replace deltaT with dTtracerLev(k) for asynchronous time stepping |
290 |
|
- global_with_exf/input.yearly verification; changes output.yearly.txt |
291 |
|
o eesupp/src: |
292 |
|
- PRINT_MESSAGE: use OpenMP critical instruction (similar to PRINT_ERROR); |
293 |
|
writing to unit 0 (if unit=errorMessageUnit) extended to the case myThid=1 |
294 |
|
- CHECK_THREADS: specific code if using OpenMP threading: improve checking of |
295 |
|
# of threads across processors (e.g., different env OMP_NUM_THREADS setting) |
296 |
|
- improve special preliminary call to EESET_PARMS (if eedata is read 2 times). |
297 |
|
- when using MPI, always set usingMPI=T ; set default accordingly and stop |
298 |
|
if not consistent. |
299 |
|
- ALL_PROC_DIE: add a BARRIER after MPI_FINALIZE call ; skip this when thread |
300 |
|
argument is zero (in case called within single-thread section). |
301 |
|
o pkg/fizhi: |
302 |
|
clean-up turbulence cold-start switch: decided in fizhi_init_vars.F, stored |
303 |
|
in common bloc (fizhi_coms.h) and then passed as argument up to S/R TURBIO. |
304 |
|
o pkg/exch2: |
305 |
|
- add argument "updateCorners" to S/R exch2_get_uv_bounds (enable to |
306 |
|
activate EXCH_IGNORE_CORNERS in vector EXCH S/R). |
307 |
|
- switch 1rst EXCH call to IGNORE_CORNERS (instead of UPDATE_CORNERS) |
308 |
|
but keep the 2nd with UPDATE_CORNERS: this prevents overwriting good data |
309 |
|
with bad ones coming from a not-yet-updated halo. |
310 |
|
This should fix the problem when sNx <> sNy and GCD(sNx,sNy) < 2*Olx. |
311 |
|
o pkg/frazil: |
312 |
|
move FrzForcT diag from frazil_diagnostics_state.F to frazil_calc_rhs.F |
313 |
|
http://mitgcm.org/pipermail/mitgcm-devel/2012-March/005384.html |
314 |
|
o model/src/config_check.F: add a check for OLx>sNx/OLy>sNy |
315 |
|
o verification/lab_sea |
316 |
|
* #undef SEAICE_EXCLUDE_FOR_EXACT_AD_TESTING |
317 |
|
* #define ALLOW_SST_CONTROL & ALLOW_SST_CONTROL |
318 |
|
* nTimeSteps 3 -> 4 |
319 |
|
* changed reference output accordingly |
320 |
|
o pkg/ctrl: |
321 |
|
* fix ctrl_get_gen for xx_sst, xx_sss |
322 |
|
o pkg/fizhi: |
323 |
|
- fix 2 calls to S/R GRD2MSC (argument igrd) in update_earth_exports.F |
324 |
|
update results of exp. fizhi-cs-32x32x40 (fail with only 6 digits for cg2d) |
325 |
|
- by default, use new version of S/R GETPWHERE (#define TRY_NEW_GETPWHERE); |
326 |
|
fix few out-of-bounds indices in referencing vars (initialise igrd to 1); |
327 |
|
clean-up fizhi_turb.F (remove unused & uninitialised qdiaglsm array). |
328 |
|
o pkg/gmredi: |
329 |
|
- add diagnostic of Redi (main diagnonal) vertical diffusive flux of Temp |
330 |
|
o pkg/fizhi: |
331 |
|
- fix few initialisation and multi-tile Pb: aqualev & gridalt now pass the |
332 |
|
(old)aces open64 tuv test. |
333 |
|
o pkg/icefront: fix bug and reorder loop in icefront_init_varia.F |
334 |
|
o pkg/mom_vecinv: |
335 |
|
- fix 1 loop range in mom_vecinv.F to be like all the other loops which |
336 |
|
set/update gu,gv/Diss (fix Pb of uninitialised fVerUkp in admom_vecinv |
337 |
|
with non-lin free-surf). |
338 |
|
o model/src: move k-loop outside of i/j-loops when (re-)computing R_low |
339 |
|
in ini/update_masks_etc.F in order to avoid -O3 optimization problems with |
340 |
|
some compilers (gfortran 4.6.0 on MacOS) |
341 |
|
o momentum: |
342 |
|
- separate fVer?(:,:,kUp) & fVer?(:,:,kDown) in argument list of MOM_FLUXFORM |
343 |
|
& MOM_VECINV subroutines (to help TAF). This fixes Pb of uninitialised |
344 |
|
fVer?(kUp) in admom_fluxform with non-lin free-surf or bottom control. |
345 |
|
|
346 |
|
checkpoint63k (2012/03/17) |
347 |
|
o model/src: |
348 |
|
- selectAddFluid (ALLOW_ADDFLUID): add missing term in horizontal momentum |
349 |
|
flux-form (mom_fluxform.F) and vertical momentum (calc_gw.F) equations. |
350 |
|
- SOLVE_DIAGONAL: switch default to use adjoinable & vectorized version |
351 |
|
(#undef SOLVE_DIAGONAL_LOWMEMORY); update output of exp. front_relax.bvp. |
352 |
|
o model/src: solve_*diagonal.F |
353 |
|
- replace division with multiplication by inverse (similar to original version) |
354 |
|
update output of AD exp. global_ocean.90x40x15. |
355 |
|
- fix pkg/gmredi/gmredi_calc_psi_bvp.F & pkg/ggl90/ggl90_calc.F |
356 |
|
to enable the use of the vectorized & differentiable solve_tridiagonal.F |
357 |
|
- test exp. global_ocean.90x40x15/input_ad: switch to advection scheme 20 for |
358 |
|
vertical advection (to get solve_tridiagonal tested in AD mode). |
359 |
|
o pkg/seaice: ECCO-CLIVAR merge: |
360 |
|
1. seaice AREA relaxation |
361 |
|
* CPP option: SEAICE_ALLOW_AREA_RELAXATION |
362 |
|
* runtime flag: SEAICE_tauAreaObsRelax |
363 |
|
2. modified global mean cost imbalance code: CPP option |
364 |
|
ALLOW_SSH_GLOBMEAN_COST_CONTRIBUTION |
365 |
|
o another iteration on solve_*diagonal.F |
366 |
|
Make choices independent of AUTODIFF |
367 |
|
Default will be the adjoinable vectorized version. |
368 |
|
Alternative choices are: |
369 |
|
* non-adjoinable original version with low memory footprint: |
370 |
|
#define SOLVE_DIAGONAL_LOWMEMORY |
371 |
|
* still adjoinable but does not vectorize (k as inner-most loop): |
372 |
|
#define SOLVE_DIAGONAL_KINNER (keeping SOLVE_DIAGONAL_LOWMEMORY undef) |
373 |
|
o testreport |
374 |
|
- add option to use a home made wrapper around "make" |
375 |
|
- remove MITGCM_IEEE env. variable (no longer active) |
376 |
|
o pkg/shelfice |
377 |
|
- improve computation of friction velocity for velocity dependent |
378 |
|
exchange coefficients: include velocities into SHELFICEBoundaryLayer |
379 |
|
block to average velocities vertically |
380 |
|
- add TAF store directives for SHI_ALLOW_GAMMAFRICT |
381 |
|
o Modify diagonal solvers, now three versions (upon request): |
382 |
|
#ifndef SOLVE_DIAGONAL_AUTODIFF |
383 |
|
1. default version (dating back a while ago) |
384 |
|
#else /* def SOLVE_DIAGONAL_AUTODIFF */ |
385 |
|
# ifdef SOLVE_DIAGONAL_KINNER |
386 |
|
2. adjoinable version with k-innermost loop (G. Forget) |
387 |
|
# else |
388 |
|
3. adjoinable vectorized version of 2. |
389 |
|
# endif |
390 |
|
#endif /* SOLVE_DIAGONAL_AUTODIFF */ |
391 |
|
All flags are currently set within routines, but could/should be moved to CPP-options. |
392 |
|
Array bounds for 1. are as before, for 2.,3. cover the entire domain |
393 |
|
(could be changed for 1. too). |
394 |
|
o verification/lab_sea (fwd): |
395 |
|
- update the 4 fwd lab_sea test experiment: no longer SEAICE_GROWTH_LEGACY, |
396 |
|
use 7 categories (except hb87), new solve4temp params (except salt_plume), |
397 |
|
constant salinity (=4) or variable salinity (salt_plume) done through |
398 |
|
SItracer. |
399 |
|
o pkg/seaice: |
400 |
|
- rename parameters: SIsalFrac to SEAICE_saltFrac & SIsal0 to SEAICE_salt0; |
401 |
|
- remove old parameters ( SEAICE_freeze, SEAICEturbFluxFormula, SEAICE_gamma_t, |
402 |
|
SEAICE_gamma_t_frz, SEAICE_availHeatTaper & SEAICE_availHeatFracFrz) |
403 |
|
from SEAICE_PARMS.h ; still need to fix seaice_cost_sst.F (broken when |
404 |
|
SEAICE_freeze got retired) but already better than previously with |
405 |
|
SEAICE_freeze=123456.7 deg.C; |
406 |
|
- move setting of facOpenGrow/Melt from SEAICE_INIT_FIXED to SEAICE_READPARMS |
407 |
|
(safer multi-threaded setting); always set SEAICEuseEVP; |
408 |
|
- setting of ocean-ice turb. flux coeff: moved from SEAICE_CHECK to |
409 |
|
SEAICE_READPARMS & SEAICE_INIT_FIXED (this fixes wrong summary report); |
410 |
|
stop if multiple specifications for the same coeff; make sure default |
411 |
|
SEAICE_mcPheePiston is compatible with drF(1) & deltaT. |
412 |
|
o pkg/obcs: |
413 |
|
- allow to switch to upwind 1rst order advection scheme for the advective |
414 |
|
flux computation at the open-boundary; disabled for adjoint built. |
415 |
|
o pkg/ptracers: |
416 |
|
- rename PTRACERS_RESTART.h -> PTRACERS_START.h |
417 |
|
- rename ptracers_fields_blocking_exchanges.F -> ptracers_fields_blocking_exch.F |
418 |
|
- add parameter to decide when to start a tracer; use internal flag to |
419 |
|
know which tracer to step-forward; switch this internal flag at the top |
420 |
|
of forward_step.F (new S/R PTRACERS_SWITCH_ONOFF); for now, not enabled |
421 |
|
for adjoint built. |
422 |
|
o pkg/seaice: |
423 |
|
- add SEAICE_SIZE.h anywhere in the code where SEAICE.h is included ; |
424 |
|
- move MULTDIM declaration and setting from SEAICE.h to SEAICE_SIZE.h ; |
425 |
|
- keep experiment offline_exf_seaice truly single category (set MULTDIM=1). |
426 |
|
o pkg/ecco: |
427 |
|
- fix writing of output file DiagnosCost_ClimSalt/Theta |
428 |
|
and DiagnosCost_DriftSalt/Theta (if multi-tiles per proc). |
429 |
|
o pkg/seaice: |
430 |
|
- add missing initialisation of COR_ICE array in dynsolver.F |
431 |
|
- merging SEAICE_MULTICATEGORY define and undef cases. Now the same code is |
432 |
|
compiled either way, and is tested in all exps (multidim was only tested |
433 |
|
in lab_sea ad before). |
434 |
|
- cosmetic change to prepare for future addition of ITD. no change to results. |
435 |
|
no defaults change. |
436 |
|
- added run time param SEAICE_multDim (7 for SEAICE_MULTICATEGORY, 1 otherwise). |
437 |
|
- by lack of adequate verification experiments, I further used custom versions |
438 |
|
of global_ocean.cs32x15 to test multicat or not, with pickup or not. All |
439 |
|
should be correct. |
440 |
|
- avoid recomputations in ad. |
441 |
|
o array initialisation in adjoint mode: |
442 |
|
1) pkg/mom_common/mom_calc_ke.F: |
443 |
|
add full array initialisation of KE inside mom_calc_ke.F (ALLOW_AUTODIFF_TAMC): |
444 |
|
The S/R which calls MOM_CALC_KE initialises KE but in some recomputation |
445 |
|
part this initialisation is gone, resulting in Floating Point Exception |
446 |
|
(caught by open64 compiler with debug option). |
447 |
|
2) model/src/convective_weights.F |
448 |
|
set output convectCount over the full array range (instead of doing |
449 |
|
an update): this fix a similar issue when called within recomputation (adjoint) |
450 |
|
which could by-pass the initialisation (normally done in the caller S/R). |
451 |
|
o pkg/seaice: |
452 |
|
- final version of turbulent flux params, as agreed at MIT ice meet: |
453 |
|
SEAICE_mcPheePiston (default is ~ 0.000875) |
454 |
|
SEAICE_mcPheeTaper (default is 0.) |
455 |
|
SEAICE_frazilFrac (defualt is 1.) |
456 |
|
SEAICE_mcPheeStepFunc (default is false) |
457 |
|
- changes results at trunction error level |
458 |
|
- ifdef SEAICE_EXTERNAL_FLUXES, disallow SEAICE_waterAlbedo |
459 |
|
o added pkg/frazil: |
460 |
|
- see frazil_description.tex for details |
461 |
|
- verification experiment is global_with_exf/input.yearly; changes output.yearly.txt |
462 |
|
o pkg/shelfice/shelfice_thermodynamics.F |
463 |
|
- fix exponents for velocity dependent exchange coefficients: change |
464 |
|
from 2/3(=0) to 0.66666666666666667 |
465 |
|
- move computation of velocity dependent exchange coefficients into |
466 |
|
separate loop, so that it applies to both thermodynamics |
467 |
|
- move some divisions by constants out of loops |
468 |
|
|
469 |
|
checkpoint63j (2012/02/16) |
470 |
|
o pkg/seaice: |
471 |
|
- overhaul of seaice_summary.F after feb 2012 clean-up of seaice_growth.F etc. |
472 |
|
- SItracer: added run time params, and allow coupling to ocn salinity. |
473 |
|
o pkg/seaice: remove (almost) all divisions by constants within loops |
474 |
|
in seaice_growth.F and seaice_solve4temp.F, this changes (only) the cg2d |
475 |
|
results of global_ocean.cs32x15.seaice from 16 to 11 digits (but only |
476 |
|
those, caused by 1/QI -> recip_QI) -> update output.seaice.txt, |
477 |
|
output_adm.seaice_dynmix.txt and also lab_sea: output_adm.evp.txt |
478 |
|
o pkg/seaice : |
479 |
|
- simplify McPhee tapering, and allow to use it with SEAICEturbFluxFormula |
480 |
|
= 1 or 2, using newly added run time parameter SEAICE_availHeatTaper. |
481 |
|
- fix d_AREAbyATM, d_AREAbyOCN, d_AREAbyICE diags I broke in r1.148. |
482 |
|
- add permanent SItflux diag that corresponds to TFLUX but includes |
483 |
|
ice+snow. Hence SItflux-TFLUX should match the ice+snow heat budget. |
484 |
|
- allow activation/testing of a fix for suspected missing term in |
485 |
|
ocn-ice heat budget (to be confirmed). To test this, you want to |
486 |
|
undef SEAICE_DISABLE_HEATCONSFIX, and then set the run time param |
487 |
|
SEAICEheatConsFix to .TRUE. that also allows the 'SIaaflux' diagnostic. |
488 |
|
- no change in results, fwd or ad. |
489 |
|
o global_ocean.cs32x15/results: 3 changes following from the pkg/seaice commit below: |
490 |
|
- switch to variable freezing point. |
491 |
|
Results change : cg2d -> 1 digits; ad -> 3-6-6 digits. |
492 |
|
- bug fix : the sublimation term that was missing is now activated. |
493 |
|
Results change : cg2d -> 2 digits; ad -> 8-6-6 digits. |
494 |
|
- bug fix : flooding sign term for SIsal0.NE.0. |
495 |
|
Result change : cg2d -> 8 digits; ad -> 11-12-9 digits. |
496 |
|
o pkg/seaice: |
497 |
|
- unified freezing point treatment : the old SEAICE_VARIABLE_FREEZING_POINT |
498 |
|
is now the default, and the old default constant freezing point is recovered |
499 |
|
with SEAICE_tempFrz0 = -1.96, |
500 |
|
SEAICE_dTempFrz_dS = 0., |
501 |
|
- retire SEAICE_freeze (= the old way of specifycing the constant freezing point) |
502 |
|
- remove ALLOW_SEAICE_FLOODING brackets; run time switch is already there. |
503 |
|
- bug fix (thanks to M. Losch) : the sublimation term that was missing is now |
504 |
|
activated. To reproduce old results with this bug define SEAICE_DISABLE_SUBLIM. |
505 |
|
- bug fix (silly me) : flooding sign term (seaice_growth.F r 1.149line 1211) |
506 |
|
was wrong for SIsal0.NE.0. Changes cs32x15 results (see above item for detail). |
507 |
|
- bug fix (thanks to O. Jahn) : area loss for melting got messed up in |
508 |
|
seaice_growth r149 for legacy branch (no results change). |
509 |
|
- introduce SEAICE_CAP_SUBLIM : caps sublimation heat flux in solve4temp (code |
510 |
|
from I. Fenty). |
511 |
|
- results did not change except for global_ocean.cs32x15, mostly due to the |
512 |
|
switch to variable freezing point (see above item for detail). |
513 |
|
o pkg/thsice: |
514 |
|
- fix and simplify interface to salt_plume pkg ; diagnose directly amount of |
515 |
|
seawater freezing in thsice_calc_thickn (not used). |
516 |
|
o pkg/seaice: turbulent flux term specification via run time params. |
517 |
|
- replace CPP switches with run time switches to facilitate testing |
518 |
|
- retired CPPs : MCPHEE_OCEAN_ICE_HEAT_FLUX, |
519 |
|
GRADIENT_MIXED_LAYER_TURBULENCE_FACTOR |
520 |
|
- added runtime parameter : SEAICEturbFluxFormula |
521 |
|
- move Mcphee constants to SEAICE_PARAMS.h |
522 |
|
- seaice_readparms.F : read turbulent flux term params (SEAICEturbFluxFormula; |
523 |
|
SEAICE_availHeatFrac/Frz, or SEAICE_gamma_t/_frz if apply); |
524 |
|
remove seaice_summary call (moved to seaice_init_fixed) |
525 |
|
- seaice_init_fixed.F : set SEAICE_availHeatFrac/Frz depending on |
526 |
|
SEAICEturbFluxFormula, and seaice_summary call (moved from seaice_readparms) |
527 |
|
- seaice_summary.F : add missing report of read turbulent flux term set up |
528 |
|
and SITRACER. |
529 |
|
- seaice_check.F : check range; inform and stop if retired CPPs are used |
530 |
|
(incl. SEAICE_AGE). |
531 |
|
- seaice_growth.F : unified use of SEAICE_availHeatFrac/Frz, without |
532 |
|
CPP switch, same computation as before for the different options. |
533 |
|
- change results at the truncation level. |
534 |
|
o pkg/seaice: |
535 |
|
- remove un-used parameters LAD & MAX_TICE ; add new run-time parameters: |
536 |
|
useMaykutSatVapPoly (default=F), postSolvTempIter (default=2) and |
537 |
|
SEAICE_wetAlbTemp (default=-1e-3). Remove option SEAICE_SOLVE4TEMP_LEGACY. |
538 |
|
- update verification/*/input*/data.seaice: |
539 |
|
to reproduce previous results with former #defined SEAICE_SOLVE4TEMP_LEGACY |
540 |
|
code, set useMaykutSatVapPoly=T & postSolvTempIter=0 ; |
541 |
|
and to reproduce previous results with former non-SOLVE4TEMP_LEGACY code, |
542 |
|
set SEAICE_wetAlbTemp=0 & SEAICE_snowThick=0. |
543 |
|
o pkg/seaice/seaice_solve4temp.F (SOLVE4TEMP_LEGACY): |
544 |
|
- remove LEGACY code for solving for tsurf (A1,A2,A3) but maintain the same |
545 |
|
algorithm (same choice: useMaykutPolySatVap=T, postSolvTempIter=0); |
546 |
|
difference in results only due to machine truncation. |
547 |
|
- remove MAX_TICE (tsurf is always =< TMELT anyway); keep MIN_TICE if using |
548 |
|
MaykutPolySatVap; keep MIN_LWDOWN. |
549 |
|
- adapt SEAICE_MODIFY_GROWTH_ADJ code (untested) to non-legacy formulation. |
550 |
|
- update output of verification exp. lab_sea (4 fwd: 11 cg2d digits except hb87, |
551 |
|
+ 3 AD: 6 digits except evp) and offline_exf_seaice.seaice (12 digits). |
552 |
|
o pkg/salt_plume: |
553 |
|
- isolate saltPlumeFlux array exch into a specific S/R called from |
554 |
|
do_oceanic_phys.F |
555 |
|
- add interface to pkg/salt_plume (filling of saltPlumeFlux) in pkg/thsice. |
556 |
|
o verification: |
557 |
|
- move untested set-up to MITgcm_contrib/verification_other. |
558 |
|
o pkg/seaice: |
559 |
|
- retire old way of SEAICE_AGE. Now replaced with ALLOW_SITRACER. |
560 |
|
o pkg/seaice: |
561 |
|
- seaice_solve4temp.F (keeping results unchanged): |
562 |
|
* add 2 future run-time params to select which saturation vapor pressure |
563 |
|
formula is used and how post-iteration fluxes are set. |
564 |
|
Temporary set these 2 params according to option SOLVE4TEMP_LEGACY; |
565 |
|
* New piece of code (not used) to update surf. atmos. heat flux F_ia using |
566 |
|
the linearized expression to stay consistent with tsurf finding. |
567 |
|
- seaice_freedrift.F avoid floating exception in sqrt(y) since y=sqrt(x*x)-x |
568 |
|
can be < 0. |
569 |
|
o pkg/seaice/seaice_solve4temp.F (SOLVE4TEMP_LEGACY undef): |
570 |
|
- remove from F_ia (and from it's derivative dFiDTs1) the contribution of |
571 |
|
conductive heat flux F_c , and add it explicitly when updating tsurf as |
572 |
|
solution of: Fc = Fia + d/dT(Fia - Fc) *delta.tsurf |
573 |
|
so that now F_ia has a consistent meaning through the entire routine. |
574 |
|
This causes differences at truncation level: update output of fwd exp. |
575 |
|
global_ocean.cs32x15.seaice. |
576 |
|
o pkg/seaice: |
577 |
|
- lots of editing/cleaning in seaice_solve4temp.F + better documentation |
578 |
|
of arrays, sign/direction of fluxes; no effect on results. |
579 |
|
- replace hard coded 273.16 (=previous value of celsius2K) by celsius2K |
580 |
|
in seaice_budget_ocean.F & seaice_solve4temp.F (SOLVE4TEMP_LEGACY part); |
581 |
|
note: this should have been done before changing celsius2K default. |
582 |
|
- to keep results unchanged, add back old value of celsius2K in verification |
583 |
|
exp. input*/data : lab_sea (AD), offline_exf_seaice (fwd+AD) & seaice_obcs |
584 |
|
- replace hard coded seawater freezing temp in seaice_solve4temp.F (part |
585 |
|
SOLVE4TEMP_LEGACY) of 271.20 (=previous value of celsius2K + default |
586 |
|
value of SEAICE_freeze) by celsius2K+SEAICE_freeze so that seawater |
587 |
|
freezing temp is consistent with value in seaice_growth.F ; this causes |
588 |
|
differences at machine truncation level: update lab_sea AD output (3/4). |
589 |
|
o pkg/seaice: |
590 |
|
- mask (in exf_init_fixed.F) atemp,aqh,evap,precip,runoff,lwflx,areamask |
591 |
|
and climsst/sss even when using seaice (to prevent calculation that might |
592 |
|
use SST on land); does not affect any verification exp. |
593 |
|
- apply mask to taux,tauy (in seaice_get_dynforcing.F); update output |
594 |
|
of lab_sea.fd & lab_sea.hb87 and global_ocean.cs32x15.seaice_dynmix. |
595 |
|
o pkg/seaice/seaice_growth.F: add mask to avoid non-zero ice thickness |
596 |
|
over land (non-legacy code) |
597 |
|
o vectorize salt_plume package: |
598 |
|
- add extra code to salt_plume_frac and salt_plume_tendency_apply_s |
599 |
|
- this code computes plumebk everywhere (also over land), thus requires |
600 |
|
code to avoid divisions by zero and catch the case of kLev=Nr. |
601 |
|
- efficiency of salt_plume_frac is still sub-optimal |
602 |
|
o tools (genmake2): |
603 |
|
- fix makefile NAME specification (option: -makefile NAME) by adding |
604 |
|
"-f $MAKEFILE" in few places where it was missing including makedepend |
605 |
|
command (contribution from Paul). |
606 |
|
|
607 |
|
checkpoint63i (2012/01/24) |
608 |
|
o model/src: |
609 |
|
- add factorized versions of some EOS code to find_rho.F, |
610 |
|
define USE_FACTORIZED_EOS to turn it on (default for TARGET_NEC_SX) |
611 |
|
o model/src: |
612 |
|
- add celsius2K to namelist PARM01; remove unused recip_rhoNil from PARAMS.h; |
613 |
|
- change default celsius2K from 273.16 to 273.15 |
614 |
|
- to keep results unchanged, add back old value of celsius2K in verification |
615 |
|
exp. input*/data : aim.5l_cs, 1D_ocean_ice_column, lab_sea (fwd), |
616 |
|
global_ocean.cs32x15(seaice) & cpl_aim+ocn(input_atm). |
617 |
|
o model/src: add a little initialisation trick to find_rho.F and find_alpha.F |
618 |
|
to make TAF generate vectorizable code |
619 |
|
o pkg/seaice/seaice_evp.F: add code (copied from CICE) that avoids underflows. |
620 |
|
Turn on by defining SEAICE_EVP_ELIMINATE_UNDERFLOWS. Most compilers |
621 |
|
have flags that do this more efficiently. |
622 |
|
o pkg/diagnostics/diagstats_calc.F |
623 |
|
- for TARGET_NEC_SX fix the treatment of the scaling factor tmpFac=scaleFact |
624 |
|
o pkg/obcs (with exch2): |
625 |
|
- derive OB indices in overlap region from OB-InsideMask (in obcs_init_fixed.F) |
626 |
|
and remove code (in obcs_readparms.F) which was based on EXCH of indices |
627 |
|
(through buffer) and did not account for exch2 (e.g., CS-grid) topology. |
628 |
|
o pkg/exf and seaice exp: |
629 |
|
- change offline_exf_seaice.seaicetd test exp. to use default DIFF1 value (=0) |
630 |
|
so that switching off area & heff advection flags does not affect results. |
631 |
|
- change default EXF_LWDOWN_WITH_EMISSIVITY to #define in EXF_OPTIONS.h |
632 |
|
- keep exp. seaice_obcs & 1D_ocean_ice_column(ad) unchanged (using #undef |
633 |
|
EXF_LWDOWN_WITH_EMISSIVITY) and, using pkg/exf default emissivities, update |
634 |
|
results of verification experiments: 1D_ocean_ice_column(fwd), |
635 |
|
global_ocean.cs32x15.seaice(fwd+ad) & seaice_dynmix(ad), lab_sea (all 8) |
636 |
|
and offline_exf_seaice (all 4). |
637 |
|
o pkg/seaice: |
638 |
|
- remove array YNEG, RIVER, TMIX (replaced by local var) from common block; |
639 |
|
remove also areaNm1 & hEffNm1 when SEAICE_GROWTH_LEGACY is undef. |
640 |
|
- fix bug: areaNm1 & hEffNm1 are needed in seaice_growth (SEAICE_GROWTH_LEGACY) |
641 |
|
but were not filled when resp. SEAICEadvArea=F & SEAICEadvHeff=F and |
642 |
|
not multi-dim advection; |
643 |
|
- fix seaice_growth.F to compile with both SEAICE_GROWTH_LEGACY and |
644 |
|
SHORTWAVE_HEATING #undef. |
645 |
|
o optim: make Makefile more robust (?) |
646 |
|
o pkg/obcs: |
647 |
|
* fix treatment of obcs_read_pickup for TAF |
648 |
|
o pkg/shelfice: |
649 |
|
* add missing initialisation of num_shifwflx |
650 |
|
* new adjoint verification exp. for isomip |
651 |
|
* change Prandtl and Schmidt numbers from generic formulations |
652 |
|
(in terms of model diffus. and visc.) to separate runtime |
653 |
|
parameters, with Holland and Jenkins (1999) default values. |
654 |
|
o pkg/exf (interpolation): |
655 |
|
- extended input field (2 rows) near the N & S poles: |
656 |
|
* fill in with the symetric value (when even Nb of data in longitude) |
657 |
|
* add average value at the poles only for scalar quantities ; for vector |
658 |
|
component interpolation, skip the averaging and keep duplicated values, |
659 |
|
unless the 2 components have same location (uvInterp=T) in which case |
660 |
|
interpolate the 2 components together using S/R EXF_INTERP_UV. |
661 |
|
- new S/R EXF_INTERP_UV to interpolate the 2 components of a vector field |
662 |
|
together: this allows to account properly for local orientation when |
663 |
|
averaging at N & S pole. |
664 |
|
- add CPP option EXF_USE_OLD_INTERP_POLE to recover old results regarding |
665 |
|
exf-interpolation near N & S poles. Undef by default. |
666 |
|
- add debug check/print on 1rst iter if debugLevel >=2. |
667 |
|
o pkg/exf: |
668 |
|
- change masking flags definition (no longer "parameter"), put them in |
669 |
|
common block and set them in S/R EXF_INIT_FIXED. |
670 |
|
- change recognized masking flag from s,u,v to c,w,s since the mask |
671 |
|
to use (maskC,W,S) depends on position on model grid and has little to |
672 |
|
do with scalar/vector difference; fix masking if using P-coordinates. |
673 |
|
- fix masking of wind-stress (was wrong if interpolated or read on A-grid) |
674 |
|
and update results of exp. global_with_exf (x2). |
675 |
|
- fix setting of zenith-angle table when SHORTWAVE_HEATING is undef. |
676 |
|
o pkg/obcs |
677 |
|
- S/R obcs_calc_stevens: fix a small bug (gammat -> gammas where appropriate) |
678 |
|
o pkg/exf (interpolation): |
679 |
|
- assume periodicity in X only if input field cover full longitude range. |
680 |
|
- improve search for lat. index (supposed to be faster, in ~log2(ny) steps, |
681 |
|
and should vectorise). |
682 |
|
- fix input lat of the 2 added row (in case we provide N.pole data). |
683 |
|
|
684 |
|
checkpoint63h (2011/12/30) |
685 |
|
o pkg/seaice: |
686 |
|
- change declaration of UVM from _RL to _RS in order to get B-grid |
687 |
|
uIce,vIce monitor when _RS = real*4 |
688 |
|
o pkg/offline: |
689 |
|
- new S/R OFFLINE_RESET_PARMS for resetting main-model params (switch off flags); |
690 |
|
- compute vertical velocity if not read from file (switch on exactConserv |
691 |
|
& staggerTimeStep); |
692 |
|
- fix output of ptracer monitor iter & time for default PTRACERS_monitorFreq. |
693 |
|
o verification: |
694 |
|
- update grid-files (in tutorial_held_suarez_cs/input and fizhi-cs-32x32x40/input) |
695 |
|
with grid-angle generated using "utils/matlab/cs_grid/cubeCalcAngle.m": |
696 |
|
fix Pb in yG definition @ 2 missing corners (previously yG was averaged); |
697 |
|
- update output of exp. global_ocean.cs32x15.viscA4 after changing |
698 |
|
grid-file angles (used here for coriolis-3d); |
699 |
|
- remove (or gzip) few files not needed to run fizhi-cs experiments. |
700 |
|
o pkg/exf (vector interpolation): |
701 |
|
- rotate vector component toward local grid-axis orientation using grid-angles |
702 |
|
(stored in GRID.h) instead of locally computed rotation angle; keep old code |
703 |
|
available with exf option EXF_USE_OLD_VEC_ROTATION (undef by default). |
704 |
|
o pkg/seaice: |
705 |
|
- allow to compile with exf options ALLOW_ATM_TEMP or ALLOW_DOWNWARD_RADIATION |
706 |
|
undefined. Fix and remove the stop when ALLOW_RUNOFF is undefined. |
707 |
|
o pkg/PKG_OPTIONS.h files: |
708 |
|
- move #include "CPP_OPTIONS.h" just after #include "PACKAGES_CONFIG.h", |
709 |
|
outside of the #ifdef ALLOW_THISPKG / #endif block. |
710 |
|
- update verification/*/code/PKG_OPTIONS.h files. |
711 |
|
o add a verification experiment (offline_exf_seaice/input.seaice) that test |
712 |
|
the B-grid LSOR solver |
713 |
|
o pkg/seaice: sort out emissivity and long wave radiation step 2: |
714 |
|
- add emissivity for snow and ice as new run-time parameters |
715 |
|
(unfortunately) that default to the corresponding exf values; this |
716 |
|
changes results in most verification experiments so that for backward |
717 |
|
compatibility these parameter have to be set explicitly to inconsistent |
718 |
|
values (ugly); |
719 |
|
- the exf-flag EXF_LWDOWN_WITH_EMISSIVITY then gets rid off the hard-wired |
720 |
|
emissivities of 0.97 associated with lwdownloc in S/R seaice_solve4temp |
721 |
|
until we agree how to handle this bug properly |
722 |
|
o pkg/exf (interpolation): |
723 |
|
- add a set of run-time param ({inputfield}_interpMethod), one for |
724 |
|
each interpolated input field, to select the interpolation method |
725 |
|
with a value of zero switching off the interpolation; |
726 |
|
- print interpolation parameters to standard output; |
727 |
|
- refine exf_check.F accordingly ; stop if USE_NO_INTERP_RUNOFF is defined |
728 |
|
(to use instead runoff_interpMethod=0 in data.exf). |
729 |
|
- simplify default setting of interpolation parameters (for all fields, |
730 |
|
set position to match model grid cell-center position, leading to trivial |
731 |
|
interpolation in case of regular Lat-Lon grid) and fix for the case of |
732 |
|
non-uniform delY. |
733 |
|
o pkg/exf/exf_radiation.F: add the factor ocean_emissivity to lwdown, for |
734 |
|
within #ifdef EXF_LWDOWN_WITH_EMISSIVITY/#endif. Unless ocean_emissivity=1. |
735 |
|
this factor is absolutely required, because the lw exitance (= out-going |
736 |
|
long wave radiation) is emissivity*stefanBoltzmann*T^4 + rho*lwdown, |
737 |
|
where the reflectivity rho = 1-emissivity for conservation reasons. Once |
738 |
|
this is established, we can remove the CPP-Flag (if we want to). |
739 |
|
o pkg/seaice: |
740 |
|
- qualify a stop statement in S/R seaice_check: EXF_READ_EVAP and |
741 |
|
ALLOW_BULKFORMULAE are only required if SEAICE_EXTERNAL_FLUXES is defined |
742 |
|
- sort out emissivity and long wave radiation, step 1: make SEAICE_emissivity |
743 |
|
what the name implies (and not emissivity*BoltzmannConstant), this means |
744 |
|
the old default value of 5.5e-8 now means |
745 |
|
5.5e-8/5.670e-8 = 0.97001763668430343479 as ocean_emissivity in |
746 |
|
S/R exf_readparms |
747 |
|
- FV-discretization of the B-grid LSOR solver, turn on with |
748 |
|
SEAICE_LSRBNEW. This new solver also involves a slight re-organization |
749 |
|
of the calling routine dynsolver, so that B and C-grid code look a |
750 |
|
little more alike. |
751 |
|
- fix index bugs in ostres.F (B-grid code) |
752 |
|
o pkg/seaice/seaice_ocean_budget.F: change (unused) code to make it look better |
753 |
|
o pkg/seaice/seaice_readparms.F: make ocean_emissivity*stefanBoltzmann the |
754 |
|
default for SEAICE_emissivity, if useEXF; requires setting a wrong |
755 |
|
parameter value in global_ocean.cs32x15/input.seaice/data.seaice for |
756 |
|
backward compatiblity |
757 |
|
o model/src: use sigmaR in calc_ivdc |
758 |
|
|
759 |
|
checkpoint63g (2011/12/16) |
760 |
|
o pkg/seaice/seaice_growth.F: |
761 |
|
fix heat conservation of the coupled ocean-seaice system. |
762 |
|
when SEAICE_HEAT_CONSERV_FIX is defined, the mismatch of |
763 |
|
solid vs liquid water treatment is resolved via QNET. |
764 |
|
o pkg/seaice/seaice_growth.F: |
765 |
|
add missing term for sublimation in saltFlux, which was breaking conservation. |
766 |
|
o pkg/seaice/seaice_model.F: |
767 |
|
Remove this patholigical case treatment in evolution branch, |
768 |
|
where it is un-necessary and breaks conservation. |
769 |
|
Update global_ocean.cs32x15/results accordingly for |
770 |
|
output.seaice.txt output_adm.seaice.txt and output_adm.seaice_dynmix.txt |
771 |
|
o model/src: |
772 |
|
useRealFreshWaterFlux: remove masking of EmPmP (using maskInC) in |
773 |
|
external_forcing_surf.F and apply mask directly in S/R SOLVE_FOR_PRESSURE. |
774 |
|
o pkg/obcs: |
775 |
|
modify Stevens boundary conditions so that S/R obcs_calc_stevens computes |
776 |
|
tendency for temperature and salinity that is then applied to update |
777 |
|
theta and salt in S/R obcs_apply_ts; this requires proper masking of all |
778 |
|
horizontal advection and diffusion contributions on the boundary (in |
779 |
|
S/Rs gad_advection, gad_calc_rhs, gmredi_rtransport, gmredi_calc_diff), and |
780 |
|
removing the mask in S/R timestep_tracer. The advantage of this approach is |
781 |
|
that the updated tracer values include contributions from the surface |
782 |
|
forcing and vertical diffusion. |
783 |
|
This modification changes exp4.stevens because of vertical diffusion. |
784 |
|
o model/src: |
785 |
|
change non-curvilinear grid setting to also work with exch2 + blank tiles: |
786 |
|
- move delX,delY to new header file (SET_GRID.h) and adjust length to 1rst |
787 |
|
face dimensions. |
788 |
|
- new S/R INI_LOCAL_GRID to define model tile-local grid (using same units |
789 |
|
as delX,delY) and called from S/R INI_CARTESIAN_GRID, INI_CYLINDER_GRID |
790 |
|
and INI_SPHERICAL_POLAR_GRID. |
791 |
|
change test-experiment global_ocean.90x40x15: |
792 |
|
- split the domain in many more tiles (9x4 tiles of 10x10 each); |
793 |
|
- switch to exch2 and remove 1 blank-tile (but only for MPI test). |
794 |
|
- update results of both test (std & dwnslp). |
795 |
|
o pkg/diagnostics/diagstat_calc.F: fix problem with non-initialised halos in |
796 |
|
tmp-fields for TARGET_NEC_SX at the cost of extra loops and less vectorisation |
797 |
|
o model/src: |
798 |
|
- take sigma-coords related code out of update_etah.F into new S/R UPDATE_ETAWS |
799 |
|
and fix initialisation when using sigma-coords. |
800 |
|
o model/src: |
801 |
|
- improve Gael's changes from Jan 14, 2011 "addition of an 'update' of hfacc etc. |
802 |
|
to their current value at the beginning of forward_step.F": add a new run-time |
803 |
|
switch "doResetHFactors" (default=false) and use it to do (or not) the resetting. |
804 |
|
- set doResetHFactors=T in the following test exp: |
805 |
|
global_ocean.90x40x15/input (rStar) & input.dwnslp (linFS), |
806 |
|
global_ocean.cs32x15/input.seaice (rStar), hs94.cs-32x32x5/input.impIGW (linFS) |
807 |
|
and tutorial_global_oce_in_p (nonLinFS in p) |
808 |
|
o pkg/ecco |
809 |
|
- added IES cost term |
810 |
|
o verification/advect_xz |
811 |
|
- scale down the depth by 1/100 (from 200 km to just 2.km) |
812 |
|
- add secondary test using divergent flow, r-star and implicit vertical advection. |
813 |
|
- add note (check_conserve_TS.txt) and matlab script to check global |
814 |
|
conservation of T & S with NonLin-FreeSurf and Adams-Bashforth. |
815 |
|
o pkg/rbcs: |
816 |
|
- add missing relaxUFile & relaxVFile in namelist (in rbcs_readparms.F). |
817 |
|
o pkg/obcs (Stevens BCs): |
818 |
|
- fix various problems in the phase speed computations of the Stevens |
819 |
|
boundary conditions: |
820 |
|
- sign error for western and southern boundaries, |
821 |
|
- time tendencies were underestimated by factor of dTracerLev and some just |
822 |
|
zero (for multidim advection schemes) -> now estimate time tendencies |
823 |
|
from difference between previous (n) time level and the one before (n-1); |
824 |
|
this requires storing a field per tracer per boundary, also in the pickup |
825 |
|
files. |
826 |
|
- replace deltaTmom in cflMer/Zon computation with dTracerLev |
827 |
|
- set default phase velocity to cflMer/Zon |
828 |
|
- update verification experiment exp4.stevens |
829 |
|
o pkg/generic_advdiff (implicit_r): |
830 |
|
- fix implicit vertical advection conservation with AB and/or NonLin-FreeSurf |
831 |
|
(remove Tr*d/dz(w) in gad_implicit_r.F and add it in gad_calc_rhs.F); |
832 |
|
- fix implicit vertical diffusion or advection with NonLin FreeSurf |
833 |
|
(was using current recip_hFacC instead of future recip_hFacC): set future |
834 |
|
recip_hFac in thermodynamics.F and pass & use it in all vertical implicit |
835 |
|
solver S/R ; |
836 |
|
- fill diagnostics of vertical advective fluxes (computed from tendency) |
837 |
|
when using implicit vert. advection; |
838 |
|
- update output of exp that uses NonLin-FreeSurf & implicit vertical diffusion: |
839 |
|
global_ocean.90x40x15 (fwd+AD), global_ocean.cs32x15 (4 fwd +4 AD), |
840 |
|
tutorial_global_oce_in_p (fwd), tutorial_tracer_adjsens (AD) |
841 |
|
and cpl_aim+ocn (used in ocean); to be fixed: bottom_ctrl_5x5. |
842 |
|
o pkg/diagnostics: |
843 |
|
- add diags for T & S tendency which goes through Adams-Bashforth. |
844 |
|
|
845 |
|
checkpoint63f (2011/11/30) |
846 |
|
o pkg/obcs/obcs_calc_stevens.F: |
847 |
|
- join some loops to reduce number of if-statements |
848 |
|
- fix a bug in code for northern and southern boundaries. |
849 |
|
o pkg/seaice/seaice_init_varia.F |
850 |
|
- undo resetting UVM=0 that has slipped in unnoticed in r1.49 and |
851 |
|
caused zero ice-velocities for the B-grid code |
852 |
|
o pkg/seaice/seaice_init_fixed.F |
853 |
|
- fix a bug in the initialisation of the second metric terms |
854 |
|
k2AtC and k2AtZ: k2AtC was set to the value of k2AtZ and k2AtZ was 0. |
855 |
|
- this affects all seaice configurations with useCurviLinearGrid = .true. |
856 |
|
and SEAICEuseDynamics = .true. (default) and SEAICEuseMetricTerms |
857 |
|
= .true. (default) |
858 |
|
- the only verification experiment affected is global_ocean.cs32x15: |
859 |
|
icedyn, seaice, and seaice_dynmix (AD) |
860 |
|
o pkg/thsice/thsice_diagnostics_state.F |
861 |
|
- only for TARGET_NEC_SX: initialise all of tmpFld, so that overlap does |
862 |
|
not contain garbage that can lead to surprises in diagstats_calc.F |
863 |
|
o model/src: |
864 |
|
- move bi,bj loops inside INTEGR_CONTINUITY ; |
865 |
|
- call INTEGR_CONTINUITY directly from forward_step.F |
866 |
|
(previously called from momentum_correction_step.F) |
867 |
|
- call UPDATE_ETAH from integr_continuity.F and update initialise_varia.F, |
868 |
|
forward_step.F and pkg/ecco/the_main_loop.F |
869 |
|
o model/src/do_oceanic_phys.F: |
870 |
|
- reset addMass to zero only if useICEFRONT ; |
871 |
|
- separated k-loop for rhoInSitu calculation ; |
872 |
|
- add 2 diagnostics for wVel*delta^k(Rho) at constant pressure (= WdRHO_P) |
873 |
|
and at constant T,S (= WdRHOdP); |
874 |
|
NOTE: change diagnostics WRHOMASS : this is now simply the Vertical transport |
875 |
|
of In-Situ Density (Anomaly) (=wVel*rho^bar-k) (previously was Vertical |
876 |
|
transport of locally-referenced potential density); |
877 |
|
to recover old diag: (WRHOMASS)_old = (WRHOMASS)_new -1/2 * (WdRHOdP) |
878 |
|
o gluemnc: |
879 |
|
- replace ncrcat with ncks when adding variables to *.glob.nc |
880 |
|
- change the way variables without record dimensions are handled so that |
881 |
|
files that have variables with and without record variables can be glued: |
882 |
|
add a record dimension with a new unique name to be able to differentiate |
883 |
|
between variable with and without a-prior record dimensions. Remove |
884 |
|
this record dimension before adding the variable to the *.glob.nc file. |
885 |
|
This procedure automatically works also for grid.*.nc files that do not |
886 |
|
have any record dimension. |
887 |
|
o verification/cpl_aim+ocn: |
888 |
|
- generate new reference output on baudelaire (gfortran, mpich2) (previously |
889 |
|
from old aces-grid cluster, ao + ifort) and update 'run_cpl_test' script. |
890 |
|
o pkg/obcs/obcs_calc_stevens.F: |
891 |
|
- add code for northern and southern boundaries. |
892 |
|
- reorder some loops to save some operations and so that drFBar can be scalar |
893 |
|
o model/src: |
894 |
|
- re-arrange extrapolation calculation in Adams-Bashforth S/R (affects machine |
895 |
|
truncation) and add diagnostics for Adams-Bashforth tendency increment. |
896 |
|
- update test-exp. AD output: |
897 |
|
1D_ocean_ice_column, bottom_ctrl_5x5, lab_sea.noseaicedyn, obcs_ctrl; |
898 |
|
and FWD output: exp4 (x3), fizhi-cs-32x32x40, fizhi-cs-aqualev20, |
899 |
|
front_relax (x3), global_ocean.90x40x15 (x2), global_ocean.cs32x15 (x5), |
900 |
|
global_ocean_ebm, ideal_2D_oce, internal_wave, isomip (x3), lab_sea (x4), |
901 |
|
matrix_example, MLAdjust (x5), natl_box (x2), tutorial_advection_in_gyre, |
902 |
|
tutorial_global_oce_in_p, tutorial_global_oce_latlon and |
903 |
|
tutorial_plume_on_slope. |
904 |
|
o pkg/diagnostics/diagstats_calc.F: add special code for TARGET_NEC_SX |
905 |
|
that vectorises. This code uses 3 extra 2D fields, F90 array assignments |
906 |
|
and F90 intrinsic routines SUM, MINVAL, MAXVAL, but since TARGET_NEC_SX |
907 |
|
does not even come with an F77 compiler anymore, this should not be |
908 |
|
a limitation |
909 |
|
|
910 |
|
checkpoint63e (2011/11/06) |
911 |
|
o pkg/seaice adjoint of LSR: |
912 |
|
- substantially rework STORE dirs. |
913 |
|
- add new flag SEAICE_LSR_ADJOINT_ITER and parameter SOLV_MAX_FIXED |
914 |
|
this flag will do SOLV_MAX_ITERS but only store SOLV_MAX_FIXED fields |
915 |
|
and only perform SOLV_MAX_FIXED adjoint iters. |
916 |
|
If flag SEAICE_LSR_ADJOINT_ITER is undef-ined, the store in the iteration |
917 |
|
always goes to same record, meaning that the last forward field will be |
918 |
|
the one used in the adjoint - an approximation. |
919 |
|
- now produces non-zero LSR-related gradients, and acceptable gradients for |
920 |
|
An's zonal (x-z) test case (to be checked in) |
921 |
|
- lab_sea: produces acceptable gradients for lab_sea with forward/backward LSR |
922 |
|
(but these are screwed up by the thermodynamics in the first place); |
923 |
|
reduce from 4 to 3 timesteps to avoid t.d. issue as temporary solution; |
924 |
|
output_adm.txt also reflects changes to seaice_lsr |
925 |
|
- global_ocean.cs32x15: |
926 |
|
* the "default" input_ad.seaice has been re-defined to always exclude dynamics |
927 |
|
(output_adm.seaice.txt was updated) |
928 |
|
* new input_ad.seaice_dynmix tests a forward LSR / backward free-drift |
929 |
|
(output_adm.seaice_dynmix.txt) |
930 |
|
o test-exp. exp4+stevens: |
931 |
|
- switch off pTracers and nonHydrostatic (restart problems with NH); update output |
932 |
|
o pkg/seaice (lsr solver): |
933 |
|
- put coriolis term into FORCEX,Y (as in seaice_evp.F; previously in rhsU,V only); |
934 |
|
update output (change truncation) of FWD test exp lab_sea & global_ocean.cs32x15 |
935 |
|
- fix SEAICEuseFREEDRIFT for new OBCS version (ndef OBCS_UVICE_OLD) |
936 |
|
- new parameter "LSR_mixIniGuess" allows (if=2,4) to mix free-drift solution |
937 |
|
into initial guess for linearised solver; use weight ~ (1/err)^2,4. |
938 |
|
- compute (with new S/R) and print initial (+ free-driff sol. if available) |
939 |
|
and final residual of linearised system. |
940 |
|
o pkg/pp81: add diagnostics for vertical eddy viscosity and diffusivity |
941 |
|
(requires new subroutines pp81_init_fixed.F/pp81_diagnostics_init.F) |
942 |
|
o pkg/obcs: Stevens BCs (still only for east and west) |
943 |
|
- fix Stevens BCs by using intermediate velocities of previous |
944 |
|
time step as "baroclinic" contribution rather velocity of one grid cell |
945 |
|
inside the boundaries; more consistent but makes the entire treatment |
946 |
|
more complicated, requires new pickup files and two new S/Rs: |
947 |
|
obcs_copy_uv_n and obcs_save_uv_n |
948 |
|
keep old version with flag OBCS_STEVENS_USE_INTERIOR_VELOCITY |
949 |
|
(undefined by default) |
950 |
|
- silently move computation of some grid factors into separate loops to |
951 |
|
save some recomputations |
952 |
|
- add a hack for non-linear free surface |
953 |
|
- add a verification experiment to exp4 that tests this code |
954 |
|
o make gluemnc/xplodemnc work with (moron) non-GNU shell tools |
955 |
|
|
956 |
|
checkpoint63d (2011/10/25) |
957 |
|
o pkg/obcs + seaice-dynamics: |
958 |
|
- new implementation calls OBCS_APPLY_UVICE (in seaice_dynsolver.F) before |
959 |
|
solving for uIce,vIce and prevent SEAICE_LSR or SEAICE_EVP to modify OB |
960 |
|
values (consistent with an implicit method). |
961 |
|
- add option OBCS_UVICE_OLD (in OBCS_OPTIONS.h, defined by default) to |
962 |
|
return to older implementation (same results as before); note that none of |
963 |
|
the "experimental" OBCS_SEAICE options are implemented in the new version. |
964 |
|
- switch test exp. seaice_obcs to use new version (update output). |
965 |
|
o pkg/seaice: |
966 |
|
- add few stop (B-grid & CS-exch, B-grid & obcs, AUTODIFF & SEAICE_maskRHS) |
967 |
|
- add parenthesis in Water-Ice drag calculations (evp & lsr); |
968 |
|
- in seaice_lsr.F, simplify and fix AA3 setting in case sNx or sNy=1; |
969 |
|
and fix case SEAICE_waterDrag=0 (but unstable if no or little seaice) |
970 |
|
- update output of test exp lab_sea that uses evp or lsr (fwd+ad) |
971 |
|
and global_ocean.cs32x15 (icedyn & seaice fwd+ad). |
972 |
|
o pkg/layers: |
973 |
|
- Include potential density as new coordinate (Thanks to David Munday) |
974 |
|
o pkg/dic: |
975 |
|
- Remove subroutine CALC_PCO2_APPROX_CO3 from carbon_chem.F and add |
976 |
|
carbonate computation/output to CALC_PCO2_APPROX |
977 |
|
o pkg/generic_advdiff |
978 |
|
- make TAF generate code that vectorizes: |
979 |
|
gad_dst3fl_adv_*.F, gad_u3c4_impl_r.F |
980 |
|
o model/src: |
981 |
|
- improve vectorization also for adjoint code of |
982 |
|
solve_tridiagonal and solve_pentadiagonal. As this requires up to |
983 |
|
5 new 3D-fields, this code is only turned on for the combination of |
984 |
|
ALLOW_AUTODIFF_TAMC and TARGET_NEC_SX. Also remove error message |
985 |
|
for singular matrix check (only for TARGET_NEC_SX). |
986 |
|
- remove hard-wiring of CPP-flag ALLOW_SOLVERS_KLOOPINSIDE to |
987 |
|
ALLOW_AUTODIFF. This flag now needs to be set outside of |
988 |
|
solve_tri/pentadiagonal.F |
989 |
|
o pkg/exf: |
990 |
|
- fix previous modif in exf_readparms.F (revision 1.66: treatment of |
991 |
|
optional namelist EXF_NML_OBCS). |
992 |
|
- add many debug-print (to trace down where pkg/cal error msg comes from) |
993 |
|
o pkg/mom_vecinv/mom_vi_hdissip.F |
994 |
|
- rearrange loops so that vectorization is improved and in particular |
995 |
|
the adjoint routine vectorizes. The code is a little longer now, but I |
996 |
|
think easier to read. |
997 |
|
Note: CPP-flag MOM_VI_ORIGINAL_VISCA4 is not tested. |
998 |
|
o pkg/seaice: |
999 |
|
- allow to specify initial seaice velocity from files (uIceFile,vIceFile) |
1000 |
|
o pkg/exch2: |
1001 |
|
- alternative code (without tile x tile matrix) for cumulated (global) sum; |
1002 |
|
used as default; keep using matrix version in adjustment.cs-32x32x1. |
1003 |
|
o tools/build_options: |
1004 |
|
- after merging some MPI and/or multi-threads specific optfiles so that |
1005 |
|
the modified default optfile support both built (using genmake2 -mpi |
1006 |
|
for the MPI built, using genmake2 -omp for OpenMP built), |
1007 |
|
move now obsolete optfiles to tools/build_options/unsupported/ |
1008 |
|
(add copy of email sent to support list in doc/optfiles_changes.txt) |
1009 |
|
|
1010 |
|
checkpoint63c (2011/09/28) |
1011 |
|
o pkg/ebm: |
1012 |
|
- fix ebm_zonalmean.F for several procs in Ydir (nPy > 1); update output of |
1013 |
|
test exp. global_ocean_ebm (affects machine truncation, 12 digits for cg2d) |
1014 |
|
o pkg/seaice (SItracer pickup) |
1015 |
|
* Simple code for pickup of SItracer fields (#define ALLOW_SITRACER) |
1016 |
|
To be done: |
1017 |
|
- thorough testing of missing fields in S/R seaice_read_pickup |
1018 |
|
- thorough matching between generic tracers and specific fields used |
1019 |
|
(e.g. volume-weighted age, area-weighted age, salinity, ...) |
1020 |
|
o pkg/grdchk: |
1021 |
|
- all procs calculate and print finite-difference gradient (for testreport) |
1022 |
|
- change description output used by testreport; |
1023 |
|
- AD testing: also compare and report Finite-difference gradient; |
1024 |
|
=> update all output_adm*.txt |
1025 |
|
o test exp. global_ocean_ebm: |
1026 |
|
- switch back useRealFreshWaterFlux (no problem anymore with run-off); |
1027 |
|
and turn on exactConserve; update (fwd) output. |
1028 |
|
o pkg/seaice (LSR): |
1029 |
|
- print to STDOUT number of solver iterations at SEAICE_monFreq interval. |
1030 |
|
- modify solver to deal cleanly with u/v switch at face edges (CS-grid): |
1031 |
|
* identical results with non-CS topology (but save up to half of EXCH calls) |
1032 |
|
* no much changes in cs-32 verification tests since no ice at face edges |
1033 |
|
(differences due to convergence criteria, since nb of iterations changed) |
1034 |
|
* expect some differences on lat-lon-cap grid. |
1035 |
|
update FWD results of global_ocean.cs32x15 test exp. (icedyn & seaice) |
1036 |
|
o pkg/exch2: |
1037 |
|
- add new CPP option to disable Cumul-Sum matrix (to save storage when using |
1038 |
|
many tiles); until alternative coding is added, just add a stop. |
1039 |
|
o pkg/kpp: |
1040 |
|
- move ptracer/salt correction to KPP diffusivity from calc_3d_diffusivity.F |
1041 |
|
to S/R KPP_CALC_DIFF_PTR (+ add tracer-index to argument list). |
1042 |
|
o optfiles: |
1043 |
|
- merged few MPI specific (*+mpi*) and multi-threads specific (*+mth) optfiles |
1044 |
|
into corresponding standard one, which now support different build: |
1045 |
|
(when using "genmake2 -mpi" for MPI built, "genmake2 -omp" for OpenMP built) |
1046 |
|
- add "darwin_amd64_gfortran" optfile (initially from Constantinos, tested by Gael) |
1047 |
|
o pkg/obcs (seaice): |
1048 |
|
- add missing EXCH for uIce,vIce after applying OBCS; |
1049 |
|
- remove S/R EXF_SWAPFFIELDS_X/Y , disable S/R EXF_SET_OBCS_X/Y by adding |
1050 |
|
argument = number of levels to S/R EXF_SWAPFFIELDS_XZ/YZ and S/R |
1051 |
|
EXF_SET_OBCS_XZ/YZ ; update obcs_exf_load.F; |
1052 |
|
- fix EXF_SET_OBCS_XZ/YZ for multi-threaded (assuming loaded arrays are shared) |
1053 |
|
- verification exp. seaice_obcs: split domain in 2 tiles ; add SIZE.h_mpi |
1054 |
|
and eedata.mth for MPI and mutlti-threaded testing. |
1055 |
|
o pkg/flt: |
1056 |
|
- make FLT EXCH working with multi-threads (and MPI + multi-threads); |
1057 |
|
- use buffer for floats IO (trajectories & profiles); size set in FLT_SIZE.h; |
1058 |
|
- fix pickup and trajectories + profiles output with multi-threads; |
1059 |
|
- add parameter to select group of fields to write along trajectory and |
1060 |
|
profile (default unchanged); |
1061 |
|
- only call PORT_RAND* function when flt_noise <> 0; |
1062 |
|
- add multi-threaded test for verification exp. flt_example. |
1063 |
|
o eesupp/src: |
1064 |
|
- fix S/R ALL_PROC_DIE for MPI+multi-threads. |
1065 |
|
- USE_OMP_THREADING: check that number of threads (from eedata) matches |
1066 |
|
actual number (from env.var. setting). |
1067 |
|
o pkg/ebm: |
1068 |
|
- fix multi-threading in ebm_area_t.F ; clean-up EBM.h header file. |
1069 |
|
- add multi-threaded test for verification exp. global_ocean_ebm |
1070 |
|
|
1071 |
|
checkpoint63b (2011/08/29) |
1072 |
|
o pkg/ebm: |
1073 |
|
- fix sign of runoff; |
1074 |
|
- update directly forcing fields (Qnet,EmPmR) to use standard |
1075 |
|
external_forcing_surf.F (remove ebm_forcing_surf.F); |
1076 |
|
- avoid loading forcing fields a 2nd time (1rst time from ini_forcing.F); |
1077 |
|
- set either tauThetaClimRelax (in data) or tauThetaZonRelax (in data.ebm) |
1078 |
|
and apply relaxation towards corresponding SST field (2-D map or ZonAver). |
1079 |
|
- update global_ocean_ebm parameters file data and update output (fwd+ad). |
1080 |
|
o verification/matrix_example: |
1081 |
|
- split domain in 8 tiles (2x4); add SIZE.h_mpi and eedata.mth for |
1082 |
|
MPI and mutlti-threaded testing. |
1083 |
|
o pkg/matrix: |
1084 |
|
- change name of S/R MATRIX_INIT to S/R MATRIX_INIT_VARIA ; |
1085 |
|
- fix for multi-tiles (add bi,bj indices to exp/impMatrixCounter); |
1086 |
|
- fix for multi-threaded (remove DATA statement); |
1087 |
|
- use standard DIFFERENT_MULTIPLE function; |
1088 |
|
- add MATRIX_OPTIONS.h |
1089 |
|
o verification/matrix_example: |
1090 |
|
To allow to split the domain in several tiles, change the domain size |
1091 |
|
from 31x31 to 32x32 by adding 1 row (on Southern side) + 1 column (on |
1092 |
|
Western side) of land points. |
1093 |
|
o (*_)write_pickup: |
1094 |
|
- fix few bugs in mypackage_read/write_pickup.F |
1095 |
|
- in MDS_WR_METAFILES calls, replace "myTime" by a local copy into |
1096 |
|
array of length 1 (was caught by compiler syntax-check). |
1097 |
|
o pkg/diagnostics (velocity potential solver): |
1098 |
|
- fix initialisation of main diagonal matrix (for overlap values) |
1099 |
|
o tools/build_options: |
1100 |
|
- towards single optfile for MPI and non-MPI built: merge few +mpi optfile |
1101 |
|
into previously non MPI one (linux_am64 pgf77,open64,pathf90,ifort11) |
1102 |
|
- linux_amd64_gfortran: merge +mpi_generic version into standard optfile: |
1103 |
|
* move IEEE flags to DEVEL section; take IEEE flags from +mpi_generic ; |
1104 |
|
* FAST case: for now, keep -O3 with ini_masks_etc.F in NOOPTFILES list; |
1105 |
|
and use "-devel" testreport option for standard testing on baudelaire. |
1106 |
|
o pkg/grdchk: |
1107 |
|
- Assume that all gradient-check position are within the same tile |
1108 |
|
corresponding to global indices (iGloTile,jGloTile) and that |
1109 |
|
positions (i,j indices or nbeg) are local to this tile; |
1110 |
|
- set tile local-indices and proc. number according to tile global-indices; |
1111 |
|
- remove grdchkwhichproc from namelist (retired param); |
1112 |
|
- send adjoint gradient to all procs so that all procs print it to STDOUT; |
1113 |
|
- switch on MPI test for global_ocean.90x40x15 AD test experiment. |
1114 |
|
o adding package bbl (Bottom Boundary Layer) |
1115 |
|
description in MITgcm/pkg/bbl/bbl_description.tex |
1116 |
|
example/test experiment in MITgcm_contrib/bbl |
1117 |
|
o adjoint verifications |
1118 |
|
- changed 5 experiments to multi-tiles |
1119 |
|
* global_ocean.90x40x15 |
1120 |
|
* obcs_ctrl |
1121 |
|
* offline_exf_seaice |
1122 |
|
* tutorial_dic_adjoffline |
1123 |
|
* tutorial_tracer_adjsens |
1124 |
|
changed reference output for those that crossed tile boundaries |
1125 |
|
(changed choices of gradient check positions) |
1126 |
|
- with same tiling, switch on MPI test for the last 4 (add SIZE.h_mpi) |
1127 |
|
- temporary bug fix in grdchk_loc.F for obcs |
1128 |
|
o genmake2 & testreport |
1129 |
|
- add target "obj" (just for OBJFILES files) to Makefile; |
1130 |
|
- add option to specify FEXTRAFLAGS on command line. |
1131 |
o pkg/mypackage: |
o pkg/mypackage: |
1132 |
- remove bi,bj arguments from argument list of S/R MYPACKAGE_CALC_RHS |
- remove bi,bj arguments from argument list of S/R MYPACKAGE_CALC_RHS |
1133 |
|
|