4 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
5 |
============================== |
============================== |
6 |
|
|
7 |
|
o AD_CONFIG.h file: |
8 |
|
- in genmake2: remove "#undef ALLOW_ECCO_OPTIMIZATION" from AD_CONFIG.h; |
9 |
|
- remove #include "AD_CONFIG.h" from all included OPTIONS files and |
10 |
|
add it explicitly in each fortran src file where it is needed. |
11 |
|
motivation: header file AD_CONFIG.h is generated/modified after "make depend" |
12 |
|
thus introducing potentially wrong dependency; therefore, better to include |
13 |
|
it only where it's really needed, and after any other included OPTIONS file |
14 |
|
(less tempting to bring conditional includes according to ALLOW_ADJOINT_RUN |
15 |
|
or ALLOW_TANGENTLINEAR_RUN). |
16 |
|
|
17 |
|
checkpoint63p (2012/07/06) |
18 |
|
o pkg/autodiff: |
19 |
|
- rename S/R INI_AUTODIFF to AUTODIFF_INIT_VARIA ; |
20 |
|
- create new header file "AUTODIFF_PARAMS.h" for pkg/autodiff parameters |
21 |
|
which are now read from new parameter file "data.autodiff" (inAdExact |
22 |
|
previously in "data", use{KPP/GMRedi/SEAICE}inAdMode previously in data.pkg) |
23 |
|
- add file "data.autodiff" in all verification/*/input_ad/ directories. |
24 |
|
- add run-time parameter to turn On/Off SEAICEuseFREEDRIFT in adjoint/forward |
25 |
|
mode (used in global_ocean.cs32x15.seaice_dynmix test exp.) |
26 |
|
o verification/tutorial_tracer_adjsens/ |
27 |
|
- add extra sub-verification which uses "SOM" advection scheme |
28 |
|
(2nd Order-Moment Advection Scheme, Prather Limiter) with |
29 |
|
#define GAD_ALLOW_TS_SOM_ADV in GAD_OPTIONS.h |
30 |
|
and in data: |
31 |
|
multiDimAdvection=.TRUE., |
32 |
|
tempAdvScheme=81, |
33 |
|
saltAdvScheme=81, |
34 |
|
o model/src: |
35 |
|
- introduce ALLOW_BALANCE_RELAX which allow the removal |
36 |
|
of the global mean of relaxation terms by setting |
37 |
|
balanceThetaClimRelax and balanceSaltClimRelax |
38 |
|
- disable balanceEmPmR and balanceQnet in the case when useSeaice. |
39 |
|
This case is now treated appropriately in seaice_growth.F |
40 |
|
o pkg/seaice: |
41 |
|
- do balanceEmPmR and balanceQnet based on the net fluxes |
42 |
|
(SItflux) that enter the ocean+ice system rather than QNET |
43 |
|
and EMPMR. Works best in real fresh water. |
44 |
|
- SIatmFW, SIatmQnet, and SItflux computations had to be moved |
45 |
|
out of the diagnotics bloc, and stored in global arrays. |
46 |
|
|
47 |
|
checkpoint63o (2012/06/29) |
48 |
|
o vertical mixing schemes: |
49 |
|
- remove unused S/R CALC_BUOYANCY (left from branch-atmos merging) |
50 |
|
- replace computation of density (in Function STATE1, opps_calc.F) by a call |
51 |
|
to S/R FIND_RHO_SCALAR |
52 |
|
- pass sigmaR to S/R PP81_CALC, MY82_CALC (but not yet used) and to GGL90_CALC |
53 |
|
and use it for N^2 calculation in ggl90_calc.F (save 2 density computations); |
54 |
|
affects output (vermix.ggl90) at truncation level. |
55 |
|
o testreport: |
56 |
|
- change adjustment of SIZE.h for mpi+mth testing: preserve enough tiles |
57 |
|
(nSx,nSy) for the number of threads (from eedata.mth) |
58 |
|
o model/src: |
59 |
|
- remove 2nd set of calls at the top of initialise_varia.F (Adjoint part) |
60 |
|
related to Non_Lin Free-Surf or Bottom-Control (but not longer necessary). |
61 |
|
o verification/tutorial_tracer_adjsens/ |
62 |
|
- compile with ALLOW_CG2D_NSA |
63 |
|
- run with nonlinFreeSurf=3 |
64 |
|
- update reference output_adm.txt |
65 |
|
o tools/genmake2: |
66 |
|
- add "-DHAVE_NETCDF" to the DEFINES list when NetCDF is available ; |
67 |
|
refine usePROFILES switching off (only when NetCDF is not available). |
68 |
|
o verification (atmospheric set-up): |
69 |
|
- change eosType='IDEALGAS' to eosType='IDEALG' to match variable declaration |
70 |
|
o pkg/profiles : refine profiles and files handling. |
71 |
|
o model/src and pkg/ecco : introduce run-time switch usePROFILES. |
72 |
|
o model/src/the_main_loop.F, forward_step.F and pkg/exf/EXF_OPTIONS.h : |
73 |
|
- changes to use model/src/the_main_loop.F rather than the ecco version |
74 |
|
in ecco type adjoint runs. |
75 |
|
o model/src: |
76 |
|
- move computation of recip of grid distance & area from INI_MASKS_ETC to INI_GRID |
77 |
|
- move setting of h0Fac[C,W,S] from INI_LINEAR_PHISURF to INI_MASKS_ETC |
78 |
|
- update_masks_etc.F: rename SMOOTH*_R4,R8 function to the corresponding type (RS,RL) |
79 |
|
o model/src: |
80 |
|
- fix S/R UPDATE_CG2D for ALLOW_DEPTH_CONTROL case (was empty before); |
81 |
|
update output of test experiment bottom_ctrl_5x5. |
82 |
|
o pkg/generic_advdiff: |
83 |
|
- implement vertical bi-harmonic diffusion (for now, only for T & S and |
84 |
|
only explicitly). |
85 |
|
o verification/bottom_ctrl_5x5/ |
86 |
|
update reference output (results changed quite some time ago, |
87 |
|
not sure when; but new reference results look ok in terms of grdchk) |
88 |
|
o verification/1D_ocean_ice_column/ |
89 |
|
For verification purposes, the exact adjoint rather than an approx. |
90 |
|
should be used (e.g., to comare against F.D. or TLM). |
91 |
|
removed autodiff_inadmode*_ad.F routines in code_ad/ |
92 |
|
Update results/ accordingly. |
93 |
|
o pkg/cal: |
94 |
|
- change cal_getdate.F so that output date is function of input time only |
95 |
|
(less ambiguous, not mixed up with input iteration number). |
96 |
|
- fix calendarDumps for non-zero phase-lag (S/R CAL_TIME2DUMP). |
97 |
|
o pkg/diagnostics: |
98 |
|
- implement calendarDumps also for stats-diags |
99 |
|
- implement calendarDumps also for snap-shot output (negative freq) |
100 |
|
- for now, ignores calendarDumps if non-zero phase-lag. |
101 |
|
|
102 |
|
checkpoint63n (2012/06/04) |
103 |
|
o model/src: |
104 |
|
- recompute rSurfW & S to ensure consistent column thickness (needed for rStar) |
105 |
|
between Sum_k(hFac*drF) and rSurf-rLow, for now only when using pkg/shelfice; |
106 |
|
- use h0FacC, W & S to compute hFac_surfC, W & S in calc_surf_dr.F (more |
107 |
|
similar to calc_r_star.F); update results of experiment internal_wave and |
108 |
|
tutorial_global_oce_in_p (affected @ truncation level). |
109 |
|
o pkg/shelfice: |
110 |
|
- add parameterisation of advective-diffusive flux into the ice shelf |
111 |
|
according to Holland and Jenkins (1999), eq.22-33, as suggested by Jan De Ryd |
112 |
|
- simply eps3/3a computation |
113 |
|
o eesupp: |
114 |
|
- add new CPP option (DISCONNECTED_TILES) to disconnect tiles (no exchange |
115 |
|
between tiles, just fill-in edges assuming locally periodic subdomain) and |
116 |
|
add corresponding exchange subroutines (EXCH0); |
117 |
|
- new CG-solver version (_EX0) for disconnected-tiles special case. |
118 |
|
o model/src (CG-solvers): |
119 |
|
- modify Single-Reduction version (CG2D_SR) in case cg2dMaxIters is reached |
120 |
|
(similar final residual output and max-Iter number as with standard version); |
121 |
|
- allows to store and use lowest-residual solution (standard & _SR versions); |
122 |
|
- cleaning up (remove old, commented out code, LETS_MAKE_JAM ...). |
123 |
|
- by default, also compile CG2D_SR version (define ALLOW_SRCG in CPP_OPTIONS.h) |
124 |
|
o eesup/src: |
125 |
|
- global_vec_sum.F cleaning: add a BARRIER ; update comments ; remove ndef |
126 |
|
ALWAYS_USE_MPI (no longer needed after fixing usingMPI settings - March 2012) |
127 |
|
o pkg/salt_plume added to lab_sea/input_ad.noseaicedyn |
128 |
|
|
129 |
|
checkpoint63m (2012/04/27) |
130 |
|
o pkg/shelfice: |
131 |
|
- account for partical cells in surface level below ice-shelf for hydrostatic |
132 |
|
pressure calculion (set uniformFreeSurfLev to F); update isomip experiments. |
133 |
|
o ad test experiments: |
134 |
|
- update and rename to "ad_optfile.local" local AD-optfile specific to a |
135 |
|
particular adjoint verification experiment (in verification/*/code_ad/) |
136 |
|
o pkg/seaice & pkg/exf: |
137 |
|
- Part two of ECCO-CLIVAR code merges. |
138 |
|
Completes merge of seaice area relax. code. |
139 |
|
Requires unified CPP option EXF_ALLOW_SEAICE_RELAX |
140 |
|
to be set in EXF_OPTIONS.h since this pkg controls its I/O |
141 |
|
(and avoids second CPP option in SEAICE_OPTIONS.h) |
142 |
|
- change default value of costIceStart & costIceEnd to a valid date. |
143 |
|
- add snowPrecip to HSNOW in seaice_growth |
144 |
|
o pkg/ptracers: |
145 |
|
- move ptracers summary and setup check into separated S/R ; add a check for |
146 |
|
valid PTRACERS_Iter0 value (error+stop or just warning). |
147 |
|
o pkg/cal (+ pkg/ecco): |
148 |
|
- avoid calling a S/R with 2 times the same arg (FWD: unless both are only used |
149 |
|
as input; and for AD, unless both are not differentiable (e.g., k index)). |
150 |
|
o model/src: |
151 |
|
- define internal flag "uniformFreeSurfLev" (=T when free-surface level-index |
152 |
|
is everywhere k=1) to speed-up some calculations (by pass kSurf reference); |
153 |
|
temporary set to usingZCoords value (to be changed to F when using shelfice). |
154 |
|
- hydrostatic pressure: implement partial cell near the surface also for |
155 |
|
Z-coordinates (for the case where uniformFreeSurfLev=F); replace test on |
156 |
|
usingZCoords with test on uniformFreeSurfLev where needed (in NonHyd code). |
157 |
|
o pkg/ctrl/ctrl_init.F: |
158 |
|
- add a formal parameter to s/r ctrl_init_rec to print |
159 |
|
xx_$(ctrl_valiable)_file for easier debugging |
160 |
|
- move call of cal_* routines into the if-block where they are really |
161 |
|
needed so that xx_$(ctrl_variable)_period can be 0 or -12 without having |
162 |
|
to specify a startdate |
163 |
|
o pkg/seaice: |
164 |
|
- update comments in SEAICE_OPTIONS.h (+ similar update of lab_sea/code and |
165 |
|
global_ocean.cs32x15/code versions). |
166 |
|
- turn on seaice Multi-category (in data.seaice, SEAICE_multDim=7) for |
167 |
|
verification experiment global_ocean.cs32x15.seaice and update results. |
168 |
|
o pkg/exf: |
169 |
|
- new S/R EXF_GETFIELD_START to return starting-time of a forcing field |
170 |
|
(according to useExfYearlyFields). |
171 |
|
o pkg/cal: |
172 |
|
- move calendar setting & summary (previously in cal_readparams.F) into new |
173 |
|
S/R CAL_INIT_FIXED to be called later in the initialisation process, once |
174 |
|
length of the simulation is set (after SET_PARMS call); This is an attempt |
175 |
|
to fix Pb when, e.g.: pkg/longstep, modifies the length of the run. |
176 |
|
- For safety: keep a record of pkg/cal parameter settings status |
177 |
|
to allow to STOP if one pkg/cal S/R is called too early. |
178 |
|
- add a stop in all basic pkg/cal subroutines if called before setting is done. |
179 |
|
- modify cal_getdate.F: stop when nothing was returned; |
180 |
|
if called with myIter=-1, simply retrun modelBaseDate (1,2 = startDate_1,_2); |
181 |
|
this allows to remove "cal.h" from seaice_read_parms.F & init_mnc_vars.F. |
182 |
|
|
183 |
|
checkpoint63l (2012/04/04) |
184 |
|
o pkg/bbl: |
185 |
|
- fix call to FIND_RHO_2D in bbl_calc_rho.F (case multi-tiles per proc); |
186 |
|
update global_with_exf output.yearly.txt file. |
187 |
|
o pkg/cal: |
188 |
|
- start to clean-up cal_set.F ; fix for the case where baseTime is non-zero. |
189 |
|
- remove 2 unsused and very short (1 line only) functions ; |
190 |
|
- remove unsused subroutines from AD-Diff list. |
191 |
|
o model/src: |
192 |
|
- start to implement clean wait to stop in initialization S/R (config_check.F, |
193 |
|
packages_check.F, ini_parms.F and pkg/obcs/obcs_check.F + pkg/mypackage) |
194 |
|
o folowing PRINT_MESSAGE changes: |
195 |
|
- remove reminder multiple print to the same file (STDOUT) when multi-threads. |
196 |
|
o pkg/bbl: |
197 |
|
- modification so it can work with z* |
198 |
|
- replace deltaT with dTtracerLev(k) for asynchronous time stepping |
199 |
|
- global_with_exf/input.yearly verification; changes output.yearly.txt |
200 |
|
o eesupp/src: |
201 |
|
- PRINT_MESSAGE: use OpenMP critical instruction (similar to PRINT_ERROR); |
202 |
|
writing to unit 0 (if unit=errorMessageUnit) extended to the case myThid=1 |
203 |
|
- CHECK_THREADS: specific code if using OpenMP threading: improve checking of |
204 |
|
# of threads across processors (e.g., different env OMP_NUM_THREADS setting) |
205 |
|
- improve special preliminary call to EESET_PARMS (if eedata is read 2 times). |
206 |
|
- when using MPI, always set usingMPI=T ; set default accordingly and stop |
207 |
|
if not consistent. |
208 |
|
- ALL_PROC_DIE: add a BARRIER after MPI_FINALIZE call ; skip this when thread |
209 |
|
argument is zero (in case called within single-thread section). |
210 |
|
o pkg/fizhi: |
211 |
|
clean-up turbulence cold-start switch: decided in fizhi_init_vars.F, stored |
212 |
|
in common bloc (fizhi_coms.h) and then passed as argument up to S/R TURBIO. |
213 |
|
o pkg/exch2: |
214 |
|
- add argument "updateCorners" to S/R exch2_get_uv_bounds (enable to |
215 |
|
activate EXCH_IGNORE_CORNERS in vector EXCH S/R). |
216 |
|
- switch 1rst EXCH call to IGNORE_CORNERS (instead of UPDATE_CORNERS) |
217 |
|
but keep the 2nd with UPDATE_CORNERS: this prevents overwriting good data |
218 |
|
with bad ones coming from a not-yet-updated halo. |
219 |
|
This should fix the problem when sNx <> sNy and GCD(sNx,sNy) < 2*Olx. |
220 |
|
o pkg/frazil: |
221 |
|
move FrzForcT diag from frazil_diagnostics_state.F to frazil_calc_rhs.F |
222 |
|
http://mitgcm.org/pipermail/mitgcm-devel/2012-March/005384.html |
223 |
|
o model/src/config_check.F: add a check for OLx>sNx/OLy>sNy |
224 |
|
o verification/lab_sea |
225 |
|
* #undef SEAICE_EXCLUDE_FOR_EXACT_AD_TESTING |
226 |
|
* #define ALLOW_SST_CONTROL & ALLOW_SST_CONTROL |
227 |
|
* nTimeSteps 3 -> 4 |
228 |
|
* changed reference output accordingly |
229 |
|
o pkg/ctrl: |
230 |
|
* fix ctrl_get_gen for xx_sst, xx_sss |
231 |
|
o pkg/fizhi: |
232 |
|
- fix 2 calls to S/R GRD2MSC (argument igrd) in update_earth_exports.F |
233 |
|
update results of exp. fizhi-cs-32x32x40 (fail with only 6 digits for cg2d) |
234 |
|
- by default, use new version of S/R GETPWHERE (#define TRY_NEW_GETPWHERE); |
235 |
|
fix few out-of-bounds indices in referencing vars (initialise igrd to 1); |
236 |
|
clean-up fizhi_turb.F (remove unused & uninitialised qdiaglsm array). |
237 |
|
o pkg/gmredi: |
238 |
|
- add diagnostic of Redi (main diagnonal) vertical diffusive flux of Temp |
239 |
|
o pkg/fizhi: |
240 |
|
- fix few initialisation and multi-tile Pb: aqualev & gridalt now pass the |
241 |
|
(old)aces open64 tuv test. |
242 |
|
o pkg/icefront: fix bug and reorder loop in icefront_init_varia.F |
243 |
|
o pkg/mom_vecinv: |
244 |
|
- fix 1 loop range in mom_vecinv.F to be like all the other loops which |
245 |
|
set/update gu,gv/Diss (fix Pb of uninitialised fVerUkp in admom_vecinv |
246 |
|
with non-lin free-surf). |
247 |
|
o model/src: move k-loop outside of i/j-loops when (re-)computing R_low |
248 |
|
in ini/update_masks_etc.F in order to avoid -O3 optimization problems with |
249 |
|
some compilers (gfortran 4.6.0 on MacOS) |
250 |
|
o momentum: |
251 |
|
- separate fVer?(:,:,kUp) & fVer?(:,:,kDown) in argument list of MOM_FLUXFORM |
252 |
|
& MOM_VECINV subroutines (to help TAF). This fixes Pb of uninitialised |
253 |
|
fVer?(kUp) in admom_fluxform with non-lin free-surf or bottom control. |
254 |
|
|
255 |
|
checkpoint63k (2012/03/17) |
256 |
|
o model/src: |
257 |
|
- selectAddFluid (ALLOW_ADDFLUID): add missing term in horizontal momentum |
258 |
|
flux-form (mom_fluxform.F) and vertical momentum (calc_gw.F) equations. |
259 |
|
- SOLVE_DIAGONAL: switch default to use adjoinable & vectorized version |
260 |
|
(#undef SOLVE_DIAGONAL_LOWMEMORY); update output of exp. front_relax.bvp. |
261 |
|
o model/src: solve_*diagonal.F |
262 |
|
- replace division with multiplication by inverse (similar to original version) |
263 |
|
update output of AD exp. global_ocean.90x40x15. |
264 |
|
- fix pkg/gmredi/gmredi_calc_psi_bvp.F & pkg/ggl90/ggl90_calc.F |
265 |
|
to enable the use of the vectorized & differentiable solve_tridiagonal.F |
266 |
|
- test exp. global_ocean.90x40x15/input_ad: switch to advection scheme 20 for |
267 |
|
vertical advection (to get solve_tridiagonal tested in AD mode). |
268 |
|
o pkg/seaice: ECCO-CLIVAR merge: |
269 |
|
1. seaice AREA relaxation |
270 |
|
* CPP option: SEAICE_ALLOW_AREA_RELAXATION |
271 |
|
* runtime flag: SEAICE_tauAreaObsRelax |
272 |
|
2. modified global mean cost imbalance code: CPP option |
273 |
|
ALLOW_SSH_GLOBMEAN_COST_CONTRIBUTION |
274 |
|
o another iteration on solve_*diagonal.F |
275 |
|
Make choices independent of AUTODIFF |
276 |
|
Default will be the adjoinable vectorized version. |
277 |
|
Alternative choices are: |
278 |
|
* non-adjoinable original version with low memory footprint: |
279 |
|
#define SOLVE_DIAGONAL_LOWMEMORY |
280 |
|
* still adjoinable but does not vectorize (k as inner-most loop): |
281 |
|
#define SOLVE_DIAGONAL_KINNER (keeping SOLVE_DIAGONAL_LOWMEMORY undef) |
282 |
|
o testreport |
283 |
|
- add option to use a home made wrapper around "make" |
284 |
|
- remove MITGCM_IEEE env. variable (no longer active) |
285 |
|
o pkg/shelfice |
286 |
|
- improve computation of friction velocity for velocity dependent |
287 |
|
exchange coefficients: include velocities into SHELFICEBoundaryLayer |
288 |
|
block to average velocities vertically |
289 |
|
- add TAF store directives for SHI_ALLOW_GAMMAFRICT |
290 |
|
o Modify diagonal solvers, now three versions (upon request): |
291 |
|
#ifndef SOLVE_DIAGONAL_AUTODIFF |
292 |
|
1. default version (dating back a while ago) |
293 |
|
#else /* def SOLVE_DIAGONAL_AUTODIFF */ |
294 |
|
# ifdef SOLVE_DIAGONAL_KINNER |
295 |
|
2. adjoinable version with k-innermost loop (G. Forget) |
296 |
|
# else |
297 |
|
3. adjoinable vectorized version of 2. |
298 |
|
# endif |
299 |
|
#endif /* SOLVE_DIAGONAL_AUTODIFF */ |
300 |
|
All flags are currently set within routines, but could/should be moved to CPP-options. |
301 |
|
Array bounds for 1. are as before, for 2.,3. cover the entire domain |
302 |
|
(could be changed for 1. too). |
303 |
|
o verification/lab_sea (fwd): |
304 |
|
- update the 4 fwd lab_sea test experiment: no longer SEAICE_GROWTH_LEGACY, |
305 |
|
use 7 categories (except hb87), new solve4temp params (except salt_plume), |
306 |
|
constant salinity (=4) or variable salinity (salt_plume) done through |
307 |
|
SItracer. |
308 |
|
o pkg/seaice: |
309 |
|
- rename parameters: SIsalFrac to SEAICE_saltFrac & SIsal0 to SEAICE_salt0; |
310 |
|
- remove old parameters ( SEAICE_freeze, SEAICEturbFluxFormula, SEAICE_gamma_t, |
311 |
|
SEAICE_gamma_t_frz, SEAICE_availHeatTaper & SEAICE_availHeatFracFrz) |
312 |
|
from SEAICE_PARMS.h ; still need to fix seaice_cost_sst.F (broken when |
313 |
|
SEAICE_freeze got retired) but already better than previously with |
314 |
|
SEAICE_freeze=123456.7 deg.C; |
315 |
|
- move setting of facOpenGrow/Melt from SEAICE_INIT_FIXED to SEAICE_READPARMS |
316 |
|
(safer multi-threaded setting); always set SEAICEuseEVP; |
317 |
|
- setting of ocean-ice turb. flux coeff: moved from SEAICE_CHECK to |
318 |
|
SEAICE_READPARMS & SEAICE_INIT_FIXED (this fixes wrong summary report); |
319 |
|
stop if multiple specifications for the same coeff; make sure default |
320 |
|
SEAICE_mcPheePiston is compatible with drF(1) & deltaT. |
321 |
|
o pkg/obcs: |
322 |
|
- allow to switch to upwind 1rst order advection scheme for the advective |
323 |
|
flux computation at the open-boundary; disabled for adjoint built. |
324 |
|
o pkg/ptracers: |
325 |
|
- rename PTRACERS_RESTART.h -> PTRACERS_START.h |
326 |
|
- rename ptracers_fields_blocking_exchanges.F -> ptracers_fields_blocking_exch.F |
327 |
|
- add parameter to decide when to start a tracer; use internal flag to |
328 |
|
know which tracer to step-forward; switch this internal flag at the top |
329 |
|
of forward_step.F (new S/R PTRACERS_SWITCH_ONOFF); for now, not enabled |
330 |
|
for adjoint built. |
331 |
|
o pkg/seaice: |
332 |
|
- add SEAICE_SIZE.h anywhere in the code where SEAICE.h is included ; |
333 |
|
- move MULTDIM declaration and setting from SEAICE.h to SEAICE_SIZE.h ; |
334 |
|
- keep experiment offline_exf_seaice truly single category (set MULTDIM=1). |
335 |
|
o pkg/ecco: |
336 |
|
- fix writing of output file DiagnosCost_ClimSalt/Theta |
337 |
|
and DiagnosCost_DriftSalt/Theta (if multi-tiles per proc). |
338 |
|
o pkg/seaice: |
339 |
|
- add missing initialisation of COR_ICE array in dynsolver.F |
340 |
|
- merging SEAICE_MULTICATEGORY define and undef cases. Now the same code is |
341 |
|
compiled either way, and is tested in all exps (multidim was only tested |
342 |
|
in lab_sea ad before). |
343 |
|
- cosmetic change to prepare for future addition of ITD. no change to results. |
344 |
|
no defaults change. |
345 |
|
- added run time param SEAICE_multDim (7 for SEAICE_MULTICATEGORY, 1 otherwise). |
346 |
|
- by lack of adequate verification experiments, I further used custom versions |
347 |
|
of global_ocean.cs32x15 to test multicat or not, with pickup or not. All |
348 |
|
should be correct. |
349 |
|
- avoid recomputations in ad. |
350 |
|
o array initialisation in adjoint mode: |
351 |
|
1) pkg/mom_common/mom_calc_ke.F: |
352 |
|
add full array initialisation of KE inside mom_calc_ke.F (ALLOW_AUTODIFF_TAMC): |
353 |
|
The S/R which calls MOM_CALC_KE initialises KE but in some recomputation |
354 |
|
part this initialisation is gone, resulting in Floating Point Exception |
355 |
|
(caught by open64 compiler with debug option). |
356 |
|
2) model/src/convective_weights.F |
357 |
|
set output convectCount over the full array range (instead of doing |
358 |
|
an update): this fix a similar issue when called within recomputation (adjoint) |
359 |
|
which could by-pass the initialisation (normally done in the caller S/R). |
360 |
|
o pkg/seaice: |
361 |
|
- final version of turbulent flux params, as agreed at MIT ice meet: |
362 |
|
SEAICE_mcPheePiston (default is ~ 0.000875) |
363 |
|
SEAICE_mcPheeTaper (default is 0.) |
364 |
|
SEAICE_frazilFrac (defualt is 1.) |
365 |
|
SEAICE_mcPheeStepFunc (default is false) |
366 |
|
- changes results at trunction error level |
367 |
|
- ifdef SEAICE_EXTERNAL_FLUXES, disallow SEAICE_waterAlbedo |
368 |
|
o added pkg/frazil: |
369 |
|
- see frazil_description.tex for details |
370 |
|
- verification experiment is global_with_exf/input.yearly; changes output.yearly.txt |
371 |
|
o pkg/shelfice/shelfice_thermodynamics.F |
372 |
|
- fix exponents for velocity dependent exchange coefficients: change |
373 |
|
from 2/3(=0) to 0.66666666666666667 |
374 |
|
- move computation of velocity dependent exchange coefficients into |
375 |
|
separate loop, so that it applies to both thermodynamics |
376 |
|
- move some divisions by constants out of loops |
377 |
|
|
378 |
|
checkpoint63j (2012/02/16) |
379 |
|
o pkg/seaice: |
380 |
|
- overhaul of seaice_summary.F after feb 2012 clean-up of seaice_growth.F etc. |
381 |
|
- SItracer: added run time params, and allow coupling to ocn salinity. |
382 |
|
o pkg/seaice: remove (almost) all divisions by constants within loops |
383 |
|
in seaice_growth.F and seaice_solve4temp.F, this changes (only) the cg2d |
384 |
|
results of global_ocean.cs32x15.seaice from 16 to 11 digits (but only |
385 |
|
those, caused by 1/QI -> recip_QI) -> update output.seaice.txt, |
386 |
|
output_adm.seaice_dynmix.txt and also lab_sea: output_adm.evp.txt |
387 |
|
o pkg/seaice : |
388 |
|
- simplify McPhee tapering, and allow to use it with SEAICEturbFluxFormula |
389 |
|
= 1 or 2, using newly added run time parameter SEAICE_availHeatTaper. |
390 |
|
- fix d_AREAbyATM, d_AREAbyOCN, d_AREAbyICE diags I broke in r1.148. |
391 |
|
- add permanent SItflux diag that corresponds to TFLUX but includes |
392 |
|
ice+snow. Hence SItflux-TFLUX should match the ice+snow heat budget. |
393 |
|
- allow activation/testing of a fix for suspected missing term in |
394 |
|
ocn-ice heat budget (to be confirmed). To test this, you want to |
395 |
|
undef SEAICE_DISABLE_HEATCONSFIX, and then set the run time param |
396 |
|
SEAICEheatConsFix to .TRUE. that also allows the 'SIaaflux' diagnostic. |
397 |
|
- no change in results, fwd or ad. |
398 |
|
o global_ocean.cs32x15/results: 3 changes following from the pkg/seaice commit below: |
399 |
|
- switch to variable freezing point. |
400 |
|
Results change : cg2d -> 1 digits; ad -> 3-6-6 digits. |
401 |
|
- bug fix : the sublimation term that was missing is now activated. |
402 |
|
Results change : cg2d -> 2 digits; ad -> 8-6-6 digits. |
403 |
|
- bug fix : flooding sign term for SIsal0.NE.0. |
404 |
|
Result change : cg2d -> 8 digits; ad -> 11-12-9 digits. |
405 |
|
o pkg/seaice: |
406 |
|
- unified freezing point treatment : the old SEAICE_VARIABLE_FREEZING_POINT |
407 |
|
is now the default, and the old default constant freezing point is recovered |
408 |
|
with SEAICE_tempFrz0 = -1.96, |
409 |
|
SEAICE_dTempFrz_dS = 0., |
410 |
|
- retire SEAICE_freeze (= the old way of specifycing the constant freezing point) |
411 |
|
- remove ALLOW_SEAICE_FLOODING brackets; run time switch is already there. |
412 |
|
- bug fix (thanks to M. Losch) : the sublimation term that was missing is now |
413 |
|
activated. To reproduce old results with this bug define SEAICE_DISABLE_SUBLIM. |
414 |
|
- bug fix (silly me) : flooding sign term (seaice_growth.F r 1.149line 1211) |
415 |
|
was wrong for SIsal0.NE.0. Changes cs32x15 results (see above item for detail). |
416 |
|
- bug fix (thanks to O. Jahn) : area loss for melting got messed up in |
417 |
|
seaice_growth r149 for legacy branch (no results change). |
418 |
|
- introduce SEAICE_CAP_SUBLIM : caps sublimation heat flux in solve4temp (code |
419 |
|
from I. Fenty). |
420 |
|
- results did not change except for global_ocean.cs32x15, mostly due to the |
421 |
|
switch to variable freezing point (see above item for detail). |
422 |
|
o pkg/thsice: |
423 |
|
- fix and simplify interface to salt_plume pkg ; diagnose directly amount of |
424 |
|
seawater freezing in thsice_calc_thickn (not used). |
425 |
|
o pkg/seaice: turbulent flux term specification via run time params. |
426 |
|
- replace CPP switches with run time switches to facilitate testing |
427 |
|
- retired CPPs : MCPHEE_OCEAN_ICE_HEAT_FLUX, |
428 |
|
GRADIENT_MIXED_LAYER_TURBULENCE_FACTOR |
429 |
|
- added runtime parameter : SEAICEturbFluxFormula |
430 |
|
- move Mcphee constants to SEAICE_PARAMS.h |
431 |
|
- seaice_readparms.F : read turbulent flux term params (SEAICEturbFluxFormula; |
432 |
|
SEAICE_availHeatFrac/Frz, or SEAICE_gamma_t/_frz if apply); |
433 |
|
remove seaice_summary call (moved to seaice_init_fixed) |
434 |
|
- seaice_init_fixed.F : set SEAICE_availHeatFrac/Frz depending on |
435 |
|
SEAICEturbFluxFormula, and seaice_summary call (moved from seaice_readparms) |
436 |
|
- seaice_summary.F : add missing report of read turbulent flux term set up |
437 |
|
and SITRACER. |
438 |
|
- seaice_check.F : check range; inform and stop if retired CPPs are used |
439 |
|
(incl. SEAICE_AGE). |
440 |
|
- seaice_growth.F : unified use of SEAICE_availHeatFrac/Frz, without |
441 |
|
CPP switch, same computation as before for the different options. |
442 |
|
- change results at the truncation level. |
443 |
|
o pkg/seaice: |
444 |
|
- remove un-used parameters LAD & MAX_TICE ; add new run-time parameters: |
445 |
|
useMaykutSatVapPoly (default=F), postSolvTempIter (default=2) and |
446 |
|
SEAICE_wetAlbTemp (default=-1e-3). Remove option SEAICE_SOLVE4TEMP_LEGACY. |
447 |
|
- update verification/*/input*/data.seaice: |
448 |
|
to reproduce previous results with former #defined SEAICE_SOLVE4TEMP_LEGACY |
449 |
|
code, set useMaykutSatVapPoly=T & postSolvTempIter=0 ; |
450 |
|
and to reproduce previous results with former non-SOLVE4TEMP_LEGACY code, |
451 |
|
set SEAICE_wetAlbTemp=0 & SEAICE_snowThick=0. |
452 |
|
o pkg/seaice/seaice_solve4temp.F (SOLVE4TEMP_LEGACY): |
453 |
|
- remove LEGACY code for solving for tsurf (A1,A2,A3) but maintain the same |
454 |
|
algorithm (same choice: useMaykutPolySatVap=T, postSolvTempIter=0); |
455 |
|
difference in results only due to machine truncation. |
456 |
|
- remove MAX_TICE (tsurf is always =< TMELT anyway); keep MIN_TICE if using |
457 |
|
MaykutPolySatVap; keep MIN_LWDOWN. |
458 |
|
- adapt SEAICE_MODIFY_GROWTH_ADJ code (untested) to non-legacy formulation. |
459 |
|
- update output of verification exp. lab_sea (4 fwd: 11 cg2d digits except hb87, |
460 |
|
+ 3 AD: 6 digits except evp) and offline_exf_seaice.seaice (12 digits). |
461 |
|
o pkg/salt_plume: |
462 |
|
- isolate saltPlumeFlux array exch into a specific S/R called from |
463 |
|
do_oceanic_phys.F |
464 |
|
- add interface to pkg/salt_plume (filling of saltPlumeFlux) in pkg/thsice. |
465 |
|
o verification: |
466 |
|
- move untested set-up to MITgcm_contrib/verification_other. |
467 |
|
o pkg/seaice: |
468 |
|
- retire old way of SEAICE_AGE. Now replaced with ALLOW_SITRACER. |
469 |
|
o pkg/seaice: |
470 |
|
- seaice_solve4temp.F (keeping results unchanged): |
471 |
|
* add 2 future run-time params to select which saturation vapor pressure |
472 |
|
formula is used and how post-iteration fluxes are set. |
473 |
|
Temporary set these 2 params according to option SOLVE4TEMP_LEGACY; |
474 |
|
* New piece of code (not used) to update surf. atmos. heat flux F_ia using |
475 |
|
the linearized expression to stay consistent with tsurf finding. |
476 |
|
- seaice_freedrift.F avoid floating exception in sqrt(y) since y=sqrt(x*x)-x |
477 |
|
can be < 0. |
478 |
|
o pkg/seaice/seaice_solve4temp.F (SOLVE4TEMP_LEGACY undef): |
479 |
|
- remove from F_ia (and from it's derivative dFiDTs1) the contribution of |
480 |
|
conductive heat flux F_c , and add it explicitly when updating tsurf as |
481 |
|
solution of: Fc = Fia + d/dT(Fia - Fc) *delta.tsurf |
482 |
|
so that now F_ia has a consistent meaning through the entire routine. |
483 |
|
This causes differences at truncation level: update output of fwd exp. |
484 |
|
global_ocean.cs32x15.seaice. |
485 |
|
o pkg/seaice: |
486 |
|
- lots of editing/cleaning in seaice_solve4temp.F + better documentation |
487 |
|
of arrays, sign/direction of fluxes; no effect on results. |
488 |
|
- replace hard coded 273.16 (=previous value of celsius2K) by celsius2K |
489 |
|
in seaice_budget_ocean.F & seaice_solve4temp.F (SOLVE4TEMP_LEGACY part); |
490 |
|
note: this should have been done before changing celsius2K default. |
491 |
|
- to keep results unchanged, add back old value of celsius2K in verification |
492 |
|
exp. input*/data : lab_sea (AD), offline_exf_seaice (fwd+AD) & seaice_obcs |
493 |
|
- replace hard coded seawater freezing temp in seaice_solve4temp.F (part |
494 |
|
SOLVE4TEMP_LEGACY) of 271.20 (=previous value of celsius2K + default |
495 |
|
value of SEAICE_freeze) by celsius2K+SEAICE_freeze so that seawater |
496 |
|
freezing temp is consistent with value in seaice_growth.F ; this causes |
497 |
|
differences at machine truncation level: update lab_sea AD output (3/4). |
498 |
|
o pkg/seaice: |
499 |
|
- mask (in exf_init_fixed.F) atemp,aqh,evap,precip,runoff,lwflx,areamask |
500 |
|
and climsst/sss even when using seaice (to prevent calculation that might |
501 |
|
use SST on land); does not affect any verification exp. |
502 |
|
- apply mask to taux,tauy (in seaice_get_dynforcing.F); update output |
503 |
|
of lab_sea.fd & lab_sea.hb87 and global_ocean.cs32x15.seaice_dynmix. |
504 |
|
o pkg/seaice/seaice_growth.F: add mask to avoid non-zero ice thickness |
505 |
|
over land (non-legacy code) |
506 |
|
o vectorize salt_plume package: |
507 |
|
- add extra code to salt_plume_frac and salt_plume_tendency_apply_s |
508 |
|
- this code computes plumebk everywhere (also over land), thus requires |
509 |
|
code to avoid divisions by zero and catch the case of kLev=Nr. |
510 |
|
- efficiency of salt_plume_frac is still sub-optimal |
511 |
|
o tools (genmake2): |
512 |
|
- fix makefile NAME specification (option: -makefile NAME) by adding |
513 |
|
"-f $MAKEFILE" in few places where it was missing including makedepend |
514 |
|
command (contribution from Paul). |
515 |
|
|
516 |
|
checkpoint63i (2012/01/24) |
517 |
|
o model/src: |
518 |
|
- add factorized versions of some EOS code to find_rho.F, |
519 |
|
define USE_FACTORIZED_EOS to turn it on (default for TARGET_NEC_SX) |
520 |
|
o model/src: |
521 |
|
- add celsius2K to namelist PARM01; remove unused recip_rhoNil from PARAMS.h; |
522 |
|
- change default celsius2K from 273.16 to 273.15 |
523 |
|
- to keep results unchanged, add back old value of celsius2K in verification |
524 |
|
exp. input*/data : aim.5l_cs, 1D_ocean_ice_column, lab_sea (fwd), |
525 |
|
global_ocean.cs32x15(seaice) & cpl_aim+ocn(input_atm). |
526 |
|
o model/src: add a little initialisation trick to find_rho.F and find_alpha.F |
527 |
|
to make TAF generate vectorizable code |
528 |
|
o pkg/seaice/seaice_evp.F: add code (copied from CICE) that avoids underflows. |
529 |
|
Turn on by defining SEAICE_EVP_ELIMINATE_UNDERFLOWS. Most compilers |
530 |
|
have flags that do this more efficiently. |
531 |
|
o pkg/diagnostics/diagstats_calc.F |
532 |
|
- for TARGET_NEC_SX fix the treatment of the scaling factor tmpFac=scaleFact |
533 |
|
o pkg/obcs (with exch2): |
534 |
|
- derive OB indices in overlap region from OB-InsideMask (in obcs_init_fixed.F) |
535 |
|
and remove code (in obcs_readparms.F) which was based on EXCH of indices |
536 |
|
(through buffer) and did not account for exch2 (e.g., CS-grid) topology. |
537 |
|
o pkg/exf and seaice exp: |
538 |
|
- change offline_exf_seaice.seaicetd test exp. to use default DIFF1 value (=0) |
539 |
|
so that switching off area & heff advection flags does not affect results. |
540 |
|
- change default EXF_LWDOWN_WITH_EMISSIVITY to #define in EXF_OPTIONS.h |
541 |
|
- keep exp. seaice_obcs & 1D_ocean_ice_column(ad) unchanged (using #undef |
542 |
|
EXF_LWDOWN_WITH_EMISSIVITY) and, using pkg/exf default emissivities, update |
543 |
|
results of verification experiments: 1D_ocean_ice_column(fwd), |
544 |
|
global_ocean.cs32x15.seaice(fwd+ad) & seaice_dynmix(ad), lab_sea (all 8) |
545 |
|
and offline_exf_seaice (all 4). |
546 |
|
o pkg/seaice: |
547 |
|
- remove array YNEG, RIVER, TMIX (replaced by local var) from common block; |
548 |
|
remove also areaNm1 & hEffNm1 when SEAICE_GROWTH_LEGACY is undef. |
549 |
|
- fix bug: areaNm1 & hEffNm1 are needed in seaice_growth (SEAICE_GROWTH_LEGACY) |
550 |
|
but were not filled when resp. SEAICEadvArea=F & SEAICEadvHeff=F and |
551 |
|
not multi-dim advection; |
552 |
|
- fix seaice_growth.F to compile with both SEAICE_GROWTH_LEGACY and |
553 |
|
SHORTWAVE_HEATING #undef. |
554 |
|
o optim: make Makefile more robust (?) |
555 |
|
o pkg/obcs: |
556 |
|
* fix treatment of obcs_read_pickup for TAF |
557 |
|
o pkg/shelfice: |
558 |
|
* add missing initialisation of num_shifwflx |
559 |
|
* new adjoint verification exp. for isomip |
560 |
|
* change Prandtl and Schmidt numbers from generic formulations |
561 |
|
(in terms of model diffus. and visc.) to separate runtime |
562 |
|
parameters, with Holland and Jenkins (1999) default values. |
563 |
|
o pkg/exf (interpolation): |
564 |
|
- extended input field (2 rows) near the N & S poles: |
565 |
|
* fill in with the symetric value (when even Nb of data in longitude) |
566 |
|
* add average value at the poles only for scalar quantities ; for vector |
567 |
|
component interpolation, skip the averaging and keep duplicated values, |
568 |
|
unless the 2 components have same location (uvInterp=T) in which case |
569 |
|
interpolate the 2 components together using S/R EXF_INTERP_UV. |
570 |
|
- new S/R EXF_INTERP_UV to interpolate the 2 components of a vector field |
571 |
|
together: this allows to account properly for local orientation when |
572 |
|
averaging at N & S pole. |
573 |
|
- add CPP option EXF_USE_OLD_INTERP_POLE to recover old results regarding |
574 |
|
exf-interpolation near N & S poles. Undef by default. |
575 |
|
- add debug check/print on 1rst iter if debugLevel >=2. |
576 |
|
o pkg/exf: |
577 |
|
- change masking flags definition (no longer "parameter"), put them in |
578 |
|
common block and set them in S/R EXF_INIT_FIXED. |
579 |
|
- change recognized masking flag from s,u,v to c,w,s since the mask |
580 |
|
to use (maskC,W,S) depends on position on model grid and has little to |
581 |
|
do with scalar/vector difference; fix masking if using P-coordinates. |
582 |
|
- fix masking of wind-stress (was wrong if interpolated or read on A-grid) |
583 |
|
and update results of exp. global_with_exf (x2). |
584 |
|
- fix setting of zenith-angle table when SHORTWAVE_HEATING is undef. |
585 |
|
o pkg/obcs |
586 |
|
- S/R obcs_calc_stevens: fix a small bug (gammat -> gammas where appropriate) |
587 |
o pkg/exf (interpolation): |
o pkg/exf (interpolation): |
588 |
- assume periodicity in X only if input field cover full longitude range. |
- assume periodicity in X only if input field cover full longitude range. |
589 |
- improve search for lat. index (supposed to be faster, in ~log2(ny) steps, |
- improve search for lat. index (supposed to be faster, in ~log2(ny) steps, |