4 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
5 |
============================== |
============================== |
6 |
|
|
7 |
|
o pkg/seaice: |
8 |
|
- add SEAICE_SIZE.h anywhere in the code where SEAICE.h is included ; |
9 |
|
- move MULTDIM declaration and setting from SEAICE.h to SEAICE_SIZE.h ; |
10 |
|
- keep experiment offline_exf_seaice truly single category (set MULTDIM=1). |
11 |
|
o pkg/ecco: |
12 |
|
- fix writing of output file DiagnosCost_ClimSalt/Theta |
13 |
|
and DiagnosCost_DriftSalt/Theta (if multi-tiles per proc). |
14 |
|
o pkg/seaice: |
15 |
|
- add missing initialisation of COR_ICE array in dynsolver.F |
16 |
|
- merging SEAICE_MULTICATEGORY define and undef cases. Now the same code is |
17 |
|
compiled either way, and is tested in all exps (multidim was only tested |
18 |
|
in lab_sea ad before). |
19 |
|
- cosmetic change to prepare for future addition of ITD. no change to results. |
20 |
|
no defaults change. |
21 |
|
- added run time param SEAICE_multDim (7 for SEAICE_MULTICATEGORY, 1 otherwise). |
22 |
|
- by lack of adequate verification experiments, I further used custom versions |
23 |
|
of global_ocean.cs32x15 to test multicat or not, with pickup or not. All |
24 |
|
should be correct. |
25 |
|
- avoid recomputations in ad. |
26 |
|
o array initialisation in adjoint mode: |
27 |
|
1) pkg/mom_common/mom_calc_ke.F: |
28 |
|
add full array initialisation of KE inside mom_calc_ke.F (ALLOW_AUTODIFF_TAMC): |
29 |
|
The S/R which calls MOM_CALC_KE initialises KE but in some recomputation |
30 |
|
part this initialisation is gone, resulting in Floating Point Exception |
31 |
|
(caught by open64 compiler with debug option). |
32 |
|
2) model/src/convective_weights.F |
33 |
|
set output convectCount over the full array range (instead of doing |
34 |
|
an update): this fix a similar issue when called within recomputation (adjoint) |
35 |
|
which could by-pass the initialisation (normally done in the caller S/R). |
36 |
|
o pkg/seaice: |
37 |
|
- final version of turbulent flux params, as agreed at MIT ice meet: |
38 |
|
SEAICE_mcPheePiston (default is ~ 0.000875) |
39 |
|
SEAICE_mcPheeTaper (default is 0.) |
40 |
|
SEAICE_frazilFrac (defualt is 1.) |
41 |
|
SEAICE_mcPheeStepFunc (default is false) |
42 |
|
- changes results at trunction error level |
43 |
|
- ifdef SEAICE_EXTERNAL_FLUXES, disallow SEAICE_waterAlbedo |
44 |
|
o added pkg/frazil: |
45 |
|
- see frazil_description.tex for details |
46 |
|
- verification experiment is global_with_exf/input.yearly; changes output.yearly.txt |
47 |
|
o pkg/shelfice/shelfice_thermodynamics.F |
48 |
|
- fix exponents for velocity dependent exchange coefficients: change |
49 |
|
from 2/3(=0) to 0.66666666666666667 |
50 |
|
- move computation of velocity dependent exchange coefficients into |
51 |
|
separate loop, so that it applies to both thermodynamics |
52 |
|
- move some divisions by constants out of loops |
53 |
|
|
54 |
|
checkpoint63j (2012/02/16) |
55 |
|
o pkg/seaice: |
56 |
|
- overhaul of seaice_summary.F after feb 2012 clean-up of seaice_growth.F etc. |
57 |
|
- SItracer: added run time params, and allow coupling to ocn salinity. |
58 |
|
o pkg/seaice: remove (almost) all divisions by constants within loops |
59 |
|
in seaice_growth.F and seaice_solve4temp.F, this changes (only) the cg2d |
60 |
|
results of global_ocean.cs32x15.seaice from 16 to 11 digits (but only |
61 |
|
those, caused by 1/QI -> recip_QI) -> update output.seaice.txt, |
62 |
|
output_adm.seaice_dynmix.txt and also lab_sea: output_adm.evp.txt |
63 |
|
o pkg/seaice : |
64 |
|
- simplify McPhee tapering, and allow to use it with SEAICEturbFluxFormula |
65 |
|
= 1 or 2, using newly added run time parameter SEAICE_availHeatTaper. |
66 |
|
- fix d_AREAbyATM, d_AREAbyOCN, d_AREAbyICE diags I broke in r1.148. |
67 |
|
- add permanent SItflux diag that corresponds to TFLUX but includes |
68 |
|
ice+snow. Hence SItflux-TFLUX should match the ice+snow heat budget. |
69 |
|
- allow activation/testing of a fix for suspected missing term in |
70 |
|
ocn-ice heat budget (to be confirmed). To test this, you want to |
71 |
|
undef SEAICE_DISABLE_HEATCONSFIX, and then set the run time param |
72 |
|
SEAICEheatConsFix to .TRUE. that also allows the 'SIaaflux' diagnostic. |
73 |
|
- no change in results, fwd or ad. |
74 |
|
o global_ocean.cs32x15/results: 3 changes following from the pkg/seaice commit below: |
75 |
|
- switch to variable freezing point. |
76 |
|
Results change : cg2d -> 1 digits; ad -> 3-6-6 digits. |
77 |
|
- bug fix : the sublimation term that was missing is now activated. |
78 |
|
Results change : cg2d -> 2 digits; ad -> 8-6-6 digits. |
79 |
|
- bug fix : flooding sign term for SIsal0.NE.0. |
80 |
|
Result change : cg2d -> 8 digits; ad -> 11-12-9 digits. |
81 |
|
o pkg/seaice: |
82 |
|
- unified freezing point treatment : the old SEAICE_VARIABLE_FREEZING_POINT |
83 |
|
is now the default, and the old default constant freezing point is recovered |
84 |
|
with SEAICE_tempFrz0 = -1.96, |
85 |
|
SEAICE_dTempFrz_dS = 0., |
86 |
|
- retire SEAICE_freeze (= the old way of specifycing the constant freezing point) |
87 |
|
- remove ALLOW_SEAICE_FLOODING brackets; run time switch is already there. |
88 |
|
- bug fix (thanks to M. Losch) : the sublimation term that was missing is now |
89 |
|
activated. To reproduce old results with this bug define SEAICE_DISABLE_SUBLIM. |
90 |
|
- bug fix (silly me) : flooding sign term (seaice_growth.F r 1.149line 1211) |
91 |
|
was wrong for SIsal0.NE.0. Changes cs32x15 results (see above item for detail). |
92 |
|
- bug fix (thanks to O. Jahn) : area loss for melting got messed up in |
93 |
|
seaice_growth r149 for legacy branch (no results change). |
94 |
|
- introduce SEAICE_CAP_SUBLIM : caps sublimation heat flux in solve4temp (code |
95 |
|
from I. Fenty). |
96 |
|
- results did not change except for global_ocean.cs32x15, mostly due to the |
97 |
|
switch to variable freezing point (see above item for detail). |
98 |
|
o pkg/thsice: |
99 |
|
- fix and simplify interface to salt_plume pkg ; diagnose directly amount of |
100 |
|
seawater freezing in thsice_calc_thickn (not used). |
101 |
|
o pkg/seaice: turbulent flux term specification via run time params. |
102 |
|
- replace CPP switches with run time switches to facilitate testing |
103 |
|
- retired CPPs : MCPHEE_OCEAN_ICE_HEAT_FLUX, |
104 |
|
GRADIENT_MIXED_LAYER_TURBULENCE_FACTOR |
105 |
|
- added runtime parameter : SEAICEturbFluxFormula |
106 |
|
- move Mcphee constants to SEAICE_PARAMS.h |
107 |
|
- seaice_readparms.F : read turbulent flux term params (SEAICEturbFluxFormula; |
108 |
|
SEAICE_availHeatFrac/Frz, or SEAICE_gamma_t/_frz if apply); |
109 |
|
remove seaice_summary call (moved to seaice_init_fixed) |
110 |
|
- seaice_init_fixed.F : set SEAICE_availHeatFrac/Frz depending on |
111 |
|
SEAICEturbFluxFormula, and seaice_summary call (moved from seaice_readparms) |
112 |
|
- seaice_summary.F : add missing report of read turbulent flux term set up |
113 |
|
and SITRACER. |
114 |
|
- seaice_check.F : check range; inform and stop if retired CPPs are used |
115 |
|
(incl. SEAICE_AGE). |
116 |
|
- seaice_growth.F : unified use of SEAICE_availHeatFrac/Frz, without |
117 |
|
CPP switch, same computation as before for the different options. |
118 |
|
- change results at the truncation level. |
119 |
|
o pkg/seaice: |
120 |
|
- remove un-used parameters LAD & MAX_TICE ; add new run-time parameters: |
121 |
|
useMaykutSatVapPoly (default=F), postSolvTempIter (default=2) and |
122 |
|
SEAICE_wetAlbTemp (default=-1e-3). Remove option SEAICE_SOLVE4TEMP_LEGACY. |
123 |
|
- update verification/*/input*/data.seaice: |
124 |
|
to reproduce previous results with former #defined SEAICE_SOLVE4TEMP_LEGACY |
125 |
|
code, set useMaykutSatVapPoly=T & postSolvTempIter=0 ; |
126 |
|
and to reproduce previous results with former non-SOLVE4TEMP_LEGACY code, |
127 |
|
set SEAICE_wetAlbTemp=0 & SEAICE_snowThick=0. |
128 |
|
o pkg/seaice/seaice_solve4temp.F (SOLVE4TEMP_LEGACY): |
129 |
|
- remove LEGACY code for solving for tsurf (A1,A2,A3) but maintain the same |
130 |
|
algorithm (same choice: useMaykutPolySatVap=T, postSolvTempIter=0); |
131 |
|
difference in results only due to machine truncation. |
132 |
|
- remove MAX_TICE (tsurf is always =< TMELT anyway); keep MIN_TICE if using |
133 |
|
MaykutPolySatVap; keep MIN_LWDOWN. |
134 |
|
- adapt SEAICE_MODIFY_GROWTH_ADJ code (untested) to non-legacy formulation. |
135 |
|
- update output of verification exp. lab_sea (4 fwd: 11 cg2d digits except hb87, |
136 |
|
+ 3 AD: 6 digits except evp) and offline_exf_seaice.seaice (12 digits). |
137 |
|
o pkg/salt_plume: |
138 |
|
- isolate saltPlumeFlux array exch into a specific S/R called from |
139 |
|
do_oceanic_phys.F |
140 |
|
- add interface to pkg/salt_plume (filling of saltPlumeFlux) in pkg/thsice. |
141 |
|
o verification: |
142 |
|
- move untested set-up to MITgcm_contrib/verification_other. |
143 |
|
o pkg/seaice: |
144 |
|
- retire old way of SEAICE_AGE. Now replaced with ALLOW_SITRACER. |
145 |
|
o pkg/seaice: |
146 |
|
- seaice_solve4temp.F (keeping results unchanged): |
147 |
|
* add 2 future run-time params to select which saturation vapor pressure |
148 |
|
formula is used and how post-iteration fluxes are set. |
149 |
|
Temporary set these 2 params according to option SOLVE4TEMP_LEGACY; |
150 |
|
* New piece of code (not used) to update surf. atmos. heat flux F_ia using |
151 |
|
the linearized expression to stay consistent with tsurf finding. |
152 |
|
- seaice_freedrift.F avoid floating exception in sqrt(y) since y=sqrt(x*x)-x |
153 |
|
can be < 0. |
154 |
|
o pkg/seaice/seaice_solve4temp.F (SOLVE4TEMP_LEGACY undef): |
155 |
|
- remove from F_ia (and from it's derivative dFiDTs1) the contribution of |
156 |
|
conductive heat flux F_c , and add it explicitly when updating tsurf as |
157 |
|
solution of: Fc = Fia + d/dT(Fia - Fc) *delta.tsurf |
158 |
|
so that now F_ia has a consistent meaning through the entire routine. |
159 |
|
This causes differences at truncation level: update output of fwd exp. |
160 |
|
global_ocean.cs32x15.seaice. |
161 |
|
o pkg/seaice: |
162 |
|
- lots of editing/cleaning in seaice_solve4temp.F + better documentation |
163 |
|
of arrays, sign/direction of fluxes; no effect on results. |
164 |
|
- replace hard coded 273.16 (=previous value of celsius2K) by celsius2K |
165 |
|
in seaice_budget_ocean.F & seaice_solve4temp.F (SOLVE4TEMP_LEGACY part); |
166 |
|
note: this should have been done before changing celsius2K default. |
167 |
|
- to keep results unchanged, add back old value of celsius2K in verification |
168 |
|
exp. input*/data : lab_sea (AD), offline_exf_seaice (fwd+AD) & seaice_obcs |
169 |
|
- replace hard coded seawater freezing temp in seaice_solve4temp.F (part |
170 |
|
SOLVE4TEMP_LEGACY) of 271.20 (=previous value of celsius2K + default |
171 |
|
value of SEAICE_freeze) by celsius2K+SEAICE_freeze so that seawater |
172 |
|
freezing temp is consistent with value in seaice_growth.F ; this causes |
173 |
|
differences at machine truncation level: update lab_sea AD output (3/4). |
174 |
|
o pkg/seaice: |
175 |
|
- mask (in exf_init_fixed.F) atemp,aqh,evap,precip,runoff,lwflx,areamask |
176 |
|
and climsst/sss even when using seaice (to prevent calculation that might |
177 |
|
use SST on land); does not affect any verification exp. |
178 |
|
- apply mask to taux,tauy (in seaice_get_dynforcing.F); update output |
179 |
|
of lab_sea.fd & lab_sea.hb87 and global_ocean.cs32x15.seaice_dynmix. |
180 |
|
o pkg/seaice/seaice_growth.F: add mask to avoid non-zero ice thickness |
181 |
|
over land (non-legacy code) |
182 |
|
o vectorize salt_plume package: |
183 |
|
- add extra code to salt_plume_frac and salt_plume_tendency_apply_s |
184 |
|
- this code computes plumebk everywhere (also over land), thus requires |
185 |
|
code to avoid divisions by zero and catch the case of kLev=Nr. |
186 |
|
- efficiency of salt_plume_frac is still sub-optimal |
187 |
|
o tools (genmake2): |
188 |
|
- fix makefile NAME specification (option: -makefile NAME) by adding |
189 |
|
"-f $MAKEFILE" in few places where it was missing including makedepend |
190 |
|
command (contribution from Paul). |
191 |
|
|
192 |
|
checkpoint63i (2012/01/24) |
193 |
|
o model/src: |
194 |
|
- add factorized versions of some EOS code to find_rho.F, |
195 |
|
define USE_FACTORIZED_EOS to turn it on (default for TARGET_NEC_SX) |
196 |
|
o model/src: |
197 |
|
- add celsius2K to namelist PARM01; remove unused recip_rhoNil from PARAMS.h; |
198 |
|
- change default celsius2K from 273.16 to 273.15 |
199 |
|
- to keep results unchanged, add back old value of celsius2K in verification |
200 |
|
exp. input*/data : aim.5l_cs, 1D_ocean_ice_column, lab_sea (fwd), |
201 |
|
global_ocean.cs32x15(seaice) & cpl_aim+ocn(input_atm). |
202 |
|
o model/src: add a little initialisation trick to find_rho.F and find_alpha.F |
203 |
|
to make TAF generate vectorizable code |
204 |
|
o pkg/seaice/seaice_evp.F: add code (copied from CICE) that avoids underflows. |
205 |
|
Turn on by defining SEAICE_EVP_ELIMINATE_UNDERFLOWS. Most compilers |
206 |
|
have flags that do this more efficiently. |
207 |
|
o pkg/diagnostics/diagstats_calc.F |
208 |
|
- for TARGET_NEC_SX fix the treatment of the scaling factor tmpFac=scaleFact |
209 |
|
o pkg/obcs (with exch2): |
210 |
|
- derive OB indices in overlap region from OB-InsideMask (in obcs_init_fixed.F) |
211 |
|
and remove code (in obcs_readparms.F) which was based on EXCH of indices |
212 |
|
(through buffer) and did not account for exch2 (e.g., CS-grid) topology. |
213 |
|
o pkg/exf and seaice exp: |
214 |
|
- change offline_exf_seaice.seaicetd test exp. to use default DIFF1 value (=0) |
215 |
|
so that switching off area & heff advection flags does not affect results. |
216 |
|
- change default EXF_LWDOWN_WITH_EMISSIVITY to #define in EXF_OPTIONS.h |
217 |
|
- keep exp. seaice_obcs & 1D_ocean_ice_column(ad) unchanged (using #undef |
218 |
|
EXF_LWDOWN_WITH_EMISSIVITY) and, using pkg/exf default emissivities, update |
219 |
|
results of verification experiments: 1D_ocean_ice_column(fwd), |
220 |
|
global_ocean.cs32x15.seaice(fwd+ad) & seaice_dynmix(ad), lab_sea (all 8) |
221 |
|
and offline_exf_seaice (all 4). |
222 |
|
o pkg/seaice: |
223 |
|
- remove array YNEG, RIVER, TMIX (replaced by local var) from common block; |
224 |
|
remove also areaNm1 & hEffNm1 when SEAICE_GROWTH_LEGACY is undef. |
225 |
|
- fix bug: areaNm1 & hEffNm1 are needed in seaice_growth (SEAICE_GROWTH_LEGACY) |
226 |
|
but were not filled when resp. SEAICEadvArea=F & SEAICEadvHeff=F and |
227 |
|
not multi-dim advection; |
228 |
|
- fix seaice_growth.F to compile with both SEAICE_GROWTH_LEGACY and |
229 |
|
SHORTWAVE_HEATING #undef. |
230 |
|
o optim: make Makefile more robust (?) |
231 |
|
o pkg/obcs: |
232 |
|
* fix treatment of obcs_read_pickup for TAF |
233 |
|
o pkg/shelfice: |
234 |
|
* add missing initialisation of num_shifwflx |
235 |
|
* new adjoint verification exp. for isomip |
236 |
|
* change Prandtl and Schmidt numbers from generic formulations |
237 |
|
(in terms of model diffus. and visc.) to separate runtime |
238 |
|
parameters, with Holland and Jenkins (1999) default values. |
239 |
|
o pkg/exf (interpolation): |
240 |
|
- extended input field (2 rows) near the N & S poles: |
241 |
|
* fill in with the symetric value (when even Nb of data in longitude) |
242 |
|
* add average value at the poles only for scalar quantities ; for vector |
243 |
|
component interpolation, skip the averaging and keep duplicated values, |
244 |
|
unless the 2 components have same location (uvInterp=T) in which case |
245 |
|
interpolate the 2 components together using S/R EXF_INTERP_UV. |
246 |
|
- new S/R EXF_INTERP_UV to interpolate the 2 components of a vector field |
247 |
|
together: this allows to account properly for local orientation when |
248 |
|
averaging at N & S pole. |
249 |
|
- add CPP option EXF_USE_OLD_INTERP_POLE to recover old results regarding |
250 |
|
exf-interpolation near N & S poles. Undef by default. |
251 |
|
- add debug check/print on 1rst iter if debugLevel >=2. |
252 |
|
o pkg/exf: |
253 |
|
- change masking flags definition (no longer "parameter"), put them in |
254 |
|
common block and set them in S/R EXF_INIT_FIXED. |
255 |
|
- change recognized masking flag from s,u,v to c,w,s since the mask |
256 |
|
to use (maskC,W,S) depends on position on model grid and has little to |
257 |
|
do with scalar/vector difference; fix masking if using P-coordinates. |
258 |
|
- fix masking of wind-stress (was wrong if interpolated or read on A-grid) |
259 |
|
and update results of exp. global_with_exf (x2). |
260 |
|
- fix setting of zenith-angle table when SHORTWAVE_HEATING is undef. |
261 |
|
o pkg/obcs |
262 |
|
- S/R obcs_calc_stevens: fix a small bug (gammat -> gammas where appropriate) |
263 |
o pkg/exf (interpolation): |
o pkg/exf (interpolation): |
264 |
- assume periodicity in X only if input field cover full longitude range. |
- assume periodicity in X only if input field cover full longitude range. |
265 |
- 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, |