| 4 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
| 5 |
============================== |
============================== |
| 6 |
|
|
| 7 |
|
checkpoint63s (2012/09/05) |
| 8 |
|
o pkg/exch2: |
| 9 |
|
- add "if usingMPI" where it was missing (in exch2_rx?_cube*.template) |
| 10 |
|
+ to remove ALWAYS_USE_MPI in source code that TAF does not see. |
| 11 |
|
- make new S/R W2_MAP_PROCS out of w2_eeboot.F to setup exch2 CommonFlag |
| 12 |
|
and tiles to process mapping. |
| 13 |
|
- re-arrange 3 header files (W2_EXCH2_PARAMS.h, _BUFFER.h & _TOPOLOGY.h) |
| 14 |
|
moving /EXCH2_FILLVAL_RX/ & /W2_EXCH2_COMMFLAG/ to W2_EXCH2_TOPOLOGY.h; |
| 15 |
|
rename "exch2_tProc" to "W2_tileProc" and move seting of e2FillValue_RX |
| 16 |
|
from w2_eeboot.F to w2_readparms.F. |
| 17 |
|
o eesupp/src: |
| 18 |
|
- set exchNeedsMemsync & exchUsesBarrier according to "usingMPI" (instead of |
| 19 |
|
ALWAYS_USE_MPI) + moved from ini_communication_patterns.F to eeset_parms.F; |
| 20 |
|
also move setting of exchCollectStatistics from exch_init.F to eeset_parms.F |
| 21 |
|
- in eedie.F and all_proc_die.F, always call MPI_FINALIZE when ALLOW_USE_MPI |
| 22 |
|
is defined (remove condition on usingMPI) to match cases where MPI_INIT |
| 23 |
|
has been called. |
| 24 |
|
- fix bug I put in previous modif of exch_sm_3d_rx.template (not much used); |
| 25 |
|
- change default (non-MPI) value of myPid (from 1 to 0) to match MPI settings |
| 26 |
|
(so that on 1 proc, get same myPid independently of usingMPI) |
| 27 |
|
- add a stop if multi-procs but usingMPI=F ; stop also if usingMPI=T and |
| 28 |
|
useCubedSphereExchange with EXCH-1 (safer). |
| 29 |
|
- add "if usingMPI" where it was missing (+ start to remove ALWAYS_USE_MPI). |
| 30 |
|
o pkg/ggl90: |
| 31 |
|
- fix bug in filling corner mask array "mskCor" (FILL_CS_CORNER_TR_RL call) |
| 32 |
|
used in ALLOW_GGL90_SMOOTH code. |
| 33 |
|
o pkg/flt: |
| 34 |
|
- fix typo in FLT_WITHOUT_Y_PERIODICITY (the effect was to ignore stopping |
| 35 |
|
floats at the Southern end of the domain). |
| 36 |
|
o eesupp/inc: |
| 37 |
|
- add a list of CPP-options which can be set by Makefile (DEF_IN_MAKEFILE.h); |
| 38 |
|
but not used (and not to be used) anywhere, just for documentation. |
| 39 |
|
o pkg/autodiff, exf, ptracers & seaice (AD & TL output): |
| 40 |
|
- comment out the very nasty "writeBinaryPrec = writeStatePrec" line, since: |
| 41 |
|
a) can damage previously written records in any other multi-record files. |
| 42 |
|
b) if needed, can easily do it cleanly, by calling S/R WRITE_REC_3D_RL/S. |
| 43 |
|
o tools/genmake2: |
| 44 |
|
- define variable MAKE within Makefile and use it in place of 'make'; |
| 45 |
|
- update AD_CONFIG.h (if needed) before building any of the executable |
| 46 |
|
(forward, adjoint or tangent-lin); this involves a 2nd invocation of make; |
| 47 |
|
and add the usual #ifndef/#define AD_CONFIG_H in AD_CONFIG.h; |
| 48 |
|
- always print usual warning comment 'this file is automatically generated' |
| 49 |
|
from script convert_cpp_cmd2defines (+ remove this warning from genmake2). |
| 50 |
|
o pkg/cheapaml: |
| 51 |
|
- updated to correspond to the code used in accepted paper (Nico Wienders). |
| 52 |
|
o pkg/exf: |
| 53 |
|
added run time switch useAtmWind to replace ALLOW_ATM_WIND cpp switch. |
| 54 |
|
ALLOW_ATM_WIND now just sets the useAtmWind default (see exf_readparms.F) |
| 55 |
|
and force defines ALLOW_BULKFORMULAE (EXF_OPTIONS.h). |
| 56 |
|
o pkg/exf, autodiff, ctrl, ecco and seaice: |
| 57 |
|
remove ALLOW_ATM_WIND brackets, or replace them with useAtmWind ones. |
| 58 |
|
o pkg/ctrl, ecco: |
| 59 |
|
allow to compile both ALLOW_U/VSTRESS_CONTROL and ALLOW_U/VWIND_CONTROL. |
| 60 |
|
Depending on useAtmWind, one is inactive, and the other is active (see |
| 61 |
|
exf_getffields.F/exf_getsurfacefluxes.F). |
| 62 |
|
o tools/genmake2: |
| 63 |
|
- improve error msg when failing in 2nd test of check_for_broken_Ff |
| 64 |
|
- fix option '-makefile' short-name (-mf according to genmake2 -help) |
| 65 |
|
- source local config file "genmake_local" just after processing the genmake2 |
| 66 |
|
command arguments (previously done before processing arg). |
| 67 |
|
o pkg/grdchk: |
| 68 |
|
- add a call to TURNOFF_MODEL_IO for Tangent-Linear run (in this case, call |
| 69 |
|
to this routine from cost_final has been dropped in g_cost_final) |
| 70 |
|
- improve printed information (more explicit msg, no longer using "print *," |
| 71 |
|
fix fcpertminus printed value). |
| 72 |
|
o model/src: |
| 73 |
|
- move call to COST_INIT_VARIA out of packages_init_variables.F directly |
| 74 |
|
into initialise_varia.F (just after calling PACKAGES_INIT_VARIABLES): |
| 75 |
|
this prevents getting cost-funct reset in adinitialise_varia recomputations. |
| 76 |
|
o pkg/dic: |
| 77 |
|
- move initialisation of "freefe" array from from dic_biotic_forcing.F to |
| 78 |
|
fe_chem.F: this prevents TAF from dropping this initialisation in TLM code. |
| 79 |
|
o pkg/grdchk: |
| 80 |
|
- grdchk_main.F : added CALL GLOBAL_SUM_INT( ierr , myThid ) |
| 81 |
|
- grdchk_getadxx.F etc. : omit I/O if ierr.EQ.0 |
| 82 |
|
o pkg/autodiff/ctrl/grdchk : |
| 83 |
|
- attempt to sort out what works and what does not, |
| 84 |
|
with regard to I/O, and to warn/stop accordingly. |
| 85 |
|
o pkg/grdchk: |
| 86 |
|
- attempt to fix grdchk in the case of useSingleCpuIO. |
| 87 |
|
- "if ( myProcId .EQ. grdchkwhichproc .AND. ierr .EQ. 0 )" |
| 88 |
|
statements were moved from grdchk_main.F and into |
| 89 |
|
grdchk_getadxx.F grdchk_getxx.F grdchk_setxx.F. |
| 90 |
|
- a global sum is applied to record numbers icvrec/layer |
| 91 |
|
out of grdchk_loc so that they be passed to I/O routines. |
| 92 |
|
o pkg/smooth: |
| 93 |
|
- fix recomputation that appeared with nlfs : use smooth_recip_hFacC, |
| 94 |
|
smooth_hFacW, smooth_hFacS (copied at init_fixed) in 3d case. |
| 95 |
|
o verification/testreport: |
| 96 |
|
- add option '-tlm' for Tangent-Linear test (-adm,-oad,-tlm are all exclusive); |
| 97 |
|
- can also use compressed (gzip) reference output file; |
| 98 |
|
- change format (in pkg/grdchk) of TLM output (+ add forward gradient) |
| 99 |
|
to be easier to collect by testreport. |
| 100 |
|
- update existing TLM output (half out-of-date by more than 3 yrs). |
| 101 |
|
- report NETCDF & LAPACK settings (from genmake2) in output dir. |
| 102 |
|
- report run-time in each individual summary (previously only for AD test); |
| 103 |
|
for TLM test, add also TAF version. |
| 104 |
|
- add daily TLM test on (old) aces cluster + weekly test on baudelaire. |
| 105 |
|
|
| 106 |
|
checkpoint63r (2012/08/15) |
| 107 |
|
o tools/genmake2: |
| 108 |
|
- fix Makefile for target ftlall (+ update files to clean after make ftlall) |
| 109 |
|
o pkg/ecco/ecco_cost_final.F : |
| 110 |
|
- fix fc print statements (to stdout and costfunction0000) so |
| 111 |
|
that the printed value is always after the global sum. |
| 112 |
|
o pkg/seaice/seaice_dynsolver.F : |
| 113 |
|
- fix ice strength assymptote at SEAICE_area_max<1 |
| 114 |
|
(no change for the standard case of SEAICE_area_max=1) |
| 115 |
|
o some cleaning: |
| 116 |
|
- pkg source files: ensure that all *.F files from each pkg included the |
| 117 |
|
corresponding ${PKG}_OPTIONS.h file (except stand-alone src file which |
| 118 |
|
only includes CPP_EEOPTIONS.h); fix S/R CTRL_SMOOTH ; update AD experiment |
| 119 |
|
code_ad/*.F src files; add new option-file in pkg/admtlm + corresponding |
| 120 |
|
includes; |
| 121 |
|
- model/src: add PACKAGES_CONFIG.h wherever ALLOW_AUTODIFF[_TAMC] is used. |
| 122 |
|
- ALLOW_DIFFKR_CONTROL: add a stop in S/R CTRL_CHECK in case ALLOW_3D_DIFFKR |
| 123 |
|
is undef; also define /DYNVARS_DIFFKR/ common block in DYNVARS.h when |
| 124 |
|
ALLOW_DIFFKR_CONTROL is defined, to allow to compile and reach this stop. |
| 125 |
|
- update option-file CPP_OPTIONS.h of all AD experiments: |
| 126 |
|
* no longer include PACKAGES_CONFIG.h ; |
| 127 |
|
* includes ECCO_CPPOPTIONS.h if PACKAGES_CONFIG.h has been included ; |
| 128 |
|
* def/undef ALLOW_3D_DIFFKR independently of ALLOW_DIFFKR_CONTROL (before |
| 129 |
|
including ECCO_CPPOPTIONS.h). |
| 130 |
|
o pkg/cost, pkg/ctrl, pkg/ecco option-file: |
| 131 |
|
- clean-up and split content of old pkg/ecco/ECCO_CPPOPTIONS.h file into |
| 132 |
|
separated pkg option-file (done according to old doc from ECCO_CPPOPTIONS.h, |
| 133 |
|
not necessary accurate). |
| 134 |
|
- add a template for option-file ECCO_CPPOPTIONS.h, mostly empty, in |
| 135 |
|
pkg/autodiff, i.e., the 1rst pkg to compile for adjoint built, with lines to |
| 136 |
|
deliberately break compilation (a remainder that it needs to be customized) |
| 137 |
|
- in pkg/ecco, includes ECCO_OPTIONS.h instead of COST_OPTIONS.h |
| 138 |
|
- rename COST_CPPOPTIONS.h to COST_OPTIONS.h |
| 139 |
|
- rename CTRL_CPPOPTIONS.h to CTRL_OPTIONS.h |
| 140 |
|
o tools/genamke2: |
| 141 |
|
- check for lapack. |
| 142 |
|
o pkg/cost and pkg/ecco : attempt to sort out some of the limiting intricacies. |
| 143 |
|
- pkg/cost.cost_final.F : allow for use of pkg/ecco and pkg/cost cost functions |
| 144 |
|
jointly. The exclusive ifdefs were removed, to add costs terms (when defined) |
| 145 |
|
from both packages. |
| 146 |
|
- the fc global sum in pkg/ecco/ecco_cost_final.F was removed accordingly. |
| 147 |
|
- for objf_atl we use either the pkg/ecco or the pkg/cost term depending on CPPs. |
| 148 |
|
- pkg/ecco and model/src : add CPP brackets around pkg/cost and pkg/ctrl includes |
| 149 |
|
of "CTRL_SIZE.h" "ctrl.h" "ctrl_dummy.h" "optim.h" / "cost.h" accordingly. |
| 150 |
|
o pkg/cost : add comments to explain special and/or untested setups. |
| 151 |
|
o pkg/exf: |
| 152 |
|
- use EXF_OPTIONS.h to set pkg/exf options except if done in ECCO_CPPOPTIONS.h |
| 153 |
|
(previously, ALLOW_AUTODIFF_TAMC was controling this switch). |
| 154 |
|
o verification/*/code_ad: |
| 155 |
|
mimimal cleaning of AD experiments ECCO_CPPOPTIONS.h option-file: |
| 156 |
|
- reccord when this file is included (define ECCO_CPPOPTIONS_H) |
| 157 |
|
- remove the most obvious obsolete features (e.g., #define ALLOW_COST not |
| 158 |
|
allowed since Oct 2003 in any option-file except in PACKAGES_CONFIG.h) |
| 159 |
|
o pkg/seaice : |
| 160 |
|
- added possibility to change ice strength/PRESS0 relationship to |
| 161 |
|
HEFF, according to the following run time parameters |
| 162 |
|
SEAICEpresPow0 :: HEFF exponent for ice strength below SEAICEpresH0 (integer, >0) |
| 163 |
|
SEAICEpresPow1 :: HEFF exponent for ice strength above SEAICEpresH0 (integer, >0) |
| 164 |
|
SEAICEpresH0 :: HEFF threshold for ice strength (m) |
| 165 |
|
o pkg/ggl90 : |
| 166 |
|
- added run time flag mxlSurfFlag to include the code that ensure |
| 167 |
|
mixing between first and second level (previously included as a comment) |
| 168 |
|
o verification/1D_ocean_ice_column/input_ad/data.autodiff : |
| 169 |
|
- add test of newly added dumpAdByRec=T option where |
| 170 |
|
we write all records to one file per variable. |
| 171 |
|
o verification/global_ocean.cs32x15/code_ad/CPP_OPTIONS.h : |
| 172 |
|
- switch to #define ALLOW_BALANCE_RELAX and ALLOW_BALANCE_FLUXES |
| 173 |
|
to check that this code is ok with taf and compiles in adjoint. |
| 174 |
|
o pkg/autodiff : |
| 175 |
|
- added run time param dumpAdByRec : ad dump format switch. |
| 176 |
|
If true then write all records to one file per variable. |
| 177 |
|
Else write one file per record (remains the default). |
| 178 |
|
- revise seaice_ad_dump.F and exf_adjoint_snapshots_ad.F accordingly. |
| 179 |
|
o pkg/seaice: |
| 180 |
|
- Added SEAICEuseTILT run time switch : if set to |
| 181 |
|
false then omit the tilt term in seaice dynamics. |
| 182 |
|
- Fix free drift calculation. |
| 183 |
|
o pkg/shelfice: |
| 184 |
|
When initializing SHELFICEDragQuadratic, set default to be |
| 185 |
|
consistent with shiCdrag in case SHELFICEuseGammaFrict is used, |
| 186 |
|
but retain option to set it independently |
| 187 |
|
(inconsistenty spotted by Veronique Dansereau) |
| 188 |
|
o eesupp/inc: |
| 189 |
|
- in EEPARAMS.h, add real-type constant for some frequently used simple number |
| 190 |
|
(0,1,2,1/2), with name explicitly referring to RS/RL type. |
| 191 |
|
o model/src: |
| 192 |
|
- move balancing of relaxation terms to new S/R balance_relax.F. |
| 193 |
|
o pkg/autodiff: |
| 194 |
|
re-write active I/O high-level (control) S/R: |
| 195 |
|
- rename S/R ACTIVE_READ/WRITE_RL to ACTIVE_READ/WRITE_3D_RL |
| 196 |
|
- add 1 more argument (useCurrentDir) to replace the "_LOC" 2nd set of S/R. |
| 197 |
|
- use standard pkg/mdsio I/O routines (MDS_READ/WRITE_SEC_XZ/YZ and |
| 198 |
|
MDS_READ/WRITE_FIELD); change loops to multi-threads. |
| 199 |
|
- update and merge active_file_control_slice.flow into active_file_control.flow |
| 200 |
|
- update the calls in lower level S/R (active_file.F, active_file_g.F and |
| 201 |
|
active_file_ad.F) keeping the argument list and function unchanged. |
| 202 |
|
- also update lower-level "_LOC" version (call same S/R with "useCurrentDir |
| 203 |
|
=.TRUE.") and remove high-level (control) _loc version (not called anymore). |
| 204 |
|
- new low-level active I/O routines, one for each type (RL/RS), for active var |
| 205 |
|
of various (XY,XZ,YZ) shape and level number (arguments: arrShape, myNr); |
| 206 |
|
fix ctrl_init_wet.F & ecco_cost_weights.F by calling the new S/R with the |
| 207 |
|
right type. |
| 208 |
|
o autodiff and RS=real*4: |
| 209 |
|
- call the right type (_RS instead of _RL) of S/R in few output S/R (pkg/exf |
| 210 |
|
ad-monitor and snapshots); disable arrType='RS' in mdsio_gl*.F ; |
| 211 |
|
- change argument type of 3-D mask argument to _RS to match type in the call |
| 212 |
|
(in seaice_cost*.F, ctrl_*_gen.F and *ctrl_bound.F); to be continued. |
| 213 |
|
o pkg/thsice: |
| 214 |
|
- re-anable balanceEmPmR when useSEAICE=T with useThSIce=T. |
| 215 |
|
- option to correct ocean fresh-water flux for imbalance of Atmos+Land |
| 216 |
|
fresh-water flux: uniform correction or proportional to local precip. |
| 217 |
|
o pkg/ctrl: |
| 218 |
|
- new header file CTRL_SIZE.h for parameters maxCtrlArr2D, maxCtrlArr3D |
| 219 |
|
- redefine CPP option ALLOW_BOTTOMDRAG_CONTROL to |
| 220 |
|
ALLOW_BOTTOMDRAG_CONTROL_NONGENERIC |
| 221 |
|
- verification/hs94.1x64x5/ tests ALLOW_GENARR{2D,3D}_CONTROL |
| 222 |
|
|
| 223 |
|
checkpoint63q (2012/07/30) |
| 224 |
|
o pkg/autodiff: |
| 225 |
|
note about monitor and snap-shot of ad-variables: |
| 226 |
|
a) need to apply ADEXCH before monitor or output of ad-vars to include |
| 227 |
|
overlap contributions (was done for adjDump but missing for monitor). |
| 228 |
|
b) since ADEXCH implies adding overlap values to interior array, machine |
| 229 |
|
truncation error is expected when adding an ADEXCH call in the middle of |
| 230 |
|
the time-stepping (although mathematically identical). |
| 231 |
|
modifs: |
| 232 |
|
- fix some incorrect type (RL/RS, UV/scalar) in ADEXCH calls and ad-vars |
| 233 |
|
declaration (adEddyPsiX,adEddyPsiY in adcommon.h). |
| 234 |
|
- add new run-time params (dumpAdVarExch & mon_AdVarExch) to select how to |
| 235 |
|
apply ADEXCH before monitor or snap-shot of ad-variables: |
| 236 |
|
=0 : no ADEXCH ; =1 : do ADEXCH ; =2 : apply ADEXCH on a local copy. |
| 237 |
|
- implement new method (copy+adexch) for snap-shot and monitor of main model |
| 238 |
|
ad-vars (initially not used by default -> output unchanged). |
| 239 |
|
- change default (set to 2) to use new method and update few output_adm.txt |
| 240 |
|
with accurate ad-monitor (global_ocean.90x40x15, tutorial_global_oce_biogeo |
| 241 |
|
and both tutorial_tracer_adjsens tests). |
| 242 |
|
- switch 2 global_ocean.cs32x15 AD-tests to AB-2 (seaice_dynmix & thsice) |
| 243 |
|
and turn off momDissip_In_AB in the 2 other (needed for stable AB-3). |
| 244 |
|
o pkg/ctrl: |
| 245 |
|
enable the use of a variable number (maxCtrlArr2D, maxCtrlArr3D) of |
| 246 |
|
generic control variables via |
| 247 |
|
#define ALLOW_GENARR2D_CONTROL and/or #define ALLOW_GENARR3D_CONTROL |
| 248 |
|
Only required customization should be S/R ctrl_map_ini_genarr.F |
| 249 |
|
(routine in repository contains example) |
| 250 |
|
o pkg/exf: |
| 251 |
|
- fix and refine checking of surface-relaxation setting when using EXF |
| 252 |
|
o verification (adjoint exp): |
| 253 |
|
- reduce length of simulation by half for experiment global_ocean.cs32x15 |
| 254 |
|
(std, seaice & seaice_dynmix) and exp tutorial_dic_adjoffline; update output. |
| 255 |
|
o pkg/mom_common (mom_u/v_bottomdrag), pkg/shelfice (shelfice_u/v_drag) |
| 256 |
|
- replace if statements KE.ne.0 with KE.gt.0 to avoid sqrt of negative numbers |
| 257 |
|
of KE that can occur with aggressive optimization |
| 258 |
|
+ put quadratic bottom drag calculation in separated loops (only done if |
| 259 |
|
using bottomDragQuadratic) |
| 260 |
|
o model/src: CG2D_NSA |
| 261 |
|
- clean-up: stop if cg2dNormaliseRHS=T and AUTODIFF |
| 262 |
|
and replace GLOBAL_SUM calls with GLOBAL_SUM_TILE. |
| 263 |
|
- in test-exp that uses CG2D_NSA (bottom_ctrl & tutorial_tracer_adjsens), |
| 264 |
|
change cg2d residual target to be expressed in Wunits (=> do not normalise |
| 265 |
|
RHS ; this fix global_max Pb with TAF); update output. |
| 266 |
|
o verification/tutorial_tracer_adjsens: |
| 267 |
|
improve parameter consistency (and get rid of warnings): |
| 268 |
|
- use deltaTFreeSurf=deltaTtracer (better with NonLin FreeSurf); |
| 269 |
|
- switch Temp Adv Scheme from 81 to 80 (no limiter since theta < 0 is physical) |
| 270 |
|
- use staggerTimeStep (for stability) in som81; use default rotation period. |
| 271 |
|
o pkg/exch2: |
| 272 |
|
- with empty facet: fix compact-format (W2_mapIO=1) definition (was previously |
| 273 |
|
reset to 0 = 1 long line in X) |
| 274 |
|
o pkg/mdsio: |
| 275 |
|
- fix reading of dimensions from meta file for large-size domain |
| 276 |
|
o model/src: |
| 277 |
|
- fix saltIsActiveTr setting for IdealGas eosType (eosType is 6 character long); |
| 278 |
|
- to refine CS-grid check, add logical flag "hasWetCSCorners": true if using |
| 279 |
|
Cubed-Sphere Exch with CS-corners inside the domain; |
| 280 |
|
make new S/R INI_GLOBAL_DOMAIN from code in ini_linear_phisurf.F to calculate |
| 281 |
|
globalArea and to set hasWetCSCorners. |
| 282 |
|
- add warning if usingCurvilinearGrid and momAdvection in flux-form (missing |
| 283 |
|
metric terms); stop if hasWetCSCorners=T and momAdvection in flux-form. |
| 284 |
|
- re-activate the stop if useCDscheme and hasWetCSCorners=T (previously if |
| 285 |
|
useCDscheme and useCubedSphereExchange). |
| 286 |
|
o pkg/seaice/seaice_growth.F : |
| 287 |
|
- compute the SEAICE_DISABLE_HEATCONSFIX term accordingly |
| 288 |
|
for the (.NOT.useRealFreshWaterFlux).OR.(nonlinFreeSurf.EQ.0) |
| 289 |
|
when the previous code was not correct. |
| 290 |
|
- in all cases diagnoze the boundary condition mismatch to SIaaflux |
| 291 |
|
- in the case of useRealFreshWaterFlux.AND.(nonlinFreeSurf.GT.0) add |
| 292 |
|
that term to QNET (essentially to set the melt/freeze water to 0degC). |
| 293 |
|
when SEAICEheatConsFix is true, so that the ocean+ice budget heat is closed. |
| 294 |
|
o verification/global_ocean.cs32x15/input.seaice : |
| 295 |
|
- add SEAICEheatConsFix in data.seaice, ensuring closed ocean+ice heat budget. |
| 296 |
|
- add diagnostics (commented out) for budget test using my matlab codes (gcmfaces). |
| 297 |
|
- update results (output.seaice.txt) accordingly. |
| 298 |
|
o verification/lab_sea/input/data.seaice : |
| 299 |
|
- remove SEAICEheatConsFix, which is now be tested |
| 300 |
|
in global_ocean.cs32x15. The presently coded fix |
| 301 |
|
is only consistent with real fresh water + nlfs. |
| 302 |
|
- update results (output.txt) accordingly. |
| 303 |
|
o pkg/salt_plume: added diagnotic for salt tendency (oceSPtnd). |
| 304 |
|
o AD_CONFIG.h file: |
| 305 |
|
- in genmake2: remove "#undef ALLOW_ECCO_OPTIMIZATION" from AD_CONFIG.h; |
| 306 |
|
- remove #include "AD_CONFIG.h" from all included OPTIONS files and |
| 307 |
|
add it explicitly in each fortran src file where it is needed. |
| 308 |
|
motivation: header file AD_CONFIG.h is generated/modified after "make depend" |
| 309 |
|
thus introducing potentially wrong dependency; therefore, better to include |
| 310 |
|
it only where it's really needed, and after any other included OPTIONS file |
| 311 |
|
(less tempting to bring conditional includes according to ALLOW_ADJOINT_RUN |
| 312 |
|
or ALLOW_TANGENTLINEAR_RUN). |
| 313 |
|
|
| 314 |
|
checkpoint63p (2012/07/06) |
| 315 |
|
o pkg/autodiff: |
| 316 |
|
- rename S/R INI_AUTODIFF to AUTODIFF_INIT_VARIA ; |
| 317 |
|
- create new header file "AUTODIFF_PARAMS.h" for pkg/autodiff parameters |
| 318 |
|
which are now read from new parameter file "data.autodiff" (inAdExact |
| 319 |
|
previously in "data", use{KPP/GMRedi/SEAICE}inAdMode previously in data.pkg) |
| 320 |
|
- add file "data.autodiff" in all verification/*/input_ad/ directories. |
| 321 |
|
- add run-time parameter to turn On/Off SEAICEuseFREEDRIFT in adjoint/forward |
| 322 |
|
mode (used in global_ocean.cs32x15.seaice_dynmix test exp.) |
| 323 |
|
o verification/tutorial_tracer_adjsens/ |
| 324 |
|
- add extra sub-verification which uses "SOM" advection scheme |
| 325 |
|
(2nd Order-Moment Advection Scheme, Prather Limiter) with |
| 326 |
|
#define GAD_ALLOW_TS_SOM_ADV in GAD_OPTIONS.h |
| 327 |
|
and in data: |
| 328 |
|
multiDimAdvection=.TRUE., |
| 329 |
|
tempAdvScheme=81, |
| 330 |
|
saltAdvScheme=81, |
| 331 |
|
o model/src: |
| 332 |
|
- introduce ALLOW_BALANCE_RELAX which allow the removal |
| 333 |
|
of the global mean of relaxation terms by setting |
| 334 |
|
balanceThetaClimRelax and balanceSaltClimRelax |
| 335 |
|
- disable balanceEmPmR and balanceQnet in the case when useSeaice. |
| 336 |
|
This case is now treated appropriately in seaice_growth.F |
| 337 |
|
o pkg/seaice: |
| 338 |
|
- do balanceEmPmR and balanceQnet based on the net fluxes |
| 339 |
|
(SItflux) that enter the ocean+ice system rather than QNET |
| 340 |
|
and EMPMR. Works best in real fresh water. |
| 341 |
|
- SIatmFW, SIatmQnet, and SItflux computations had to be moved |
| 342 |
|
out of the diagnotics bloc, and stored in global arrays. |
| 343 |
|
|
| 344 |
|
checkpoint63o (2012/06/29) |
| 345 |
|
o vertical mixing schemes: |
| 346 |
|
- remove unused S/R CALC_BUOYANCY (left from branch-atmos merging) |
| 347 |
|
- replace computation of density (in Function STATE1, opps_calc.F) by a call |
| 348 |
|
to S/R FIND_RHO_SCALAR |
| 349 |
|
- pass sigmaR to S/R PP81_CALC, MY82_CALC (but not yet used) and to GGL90_CALC |
| 350 |
|
and use it for N^2 calculation in ggl90_calc.F (save 2 density computations); |
| 351 |
|
affects output (vermix.ggl90) at truncation level. |
| 352 |
|
o testreport: |
| 353 |
|
- change adjustment of SIZE.h for mpi+mth testing: preserve enough tiles |
| 354 |
|
(nSx,nSy) for the number of threads (from eedata.mth) |
| 355 |
|
o model/src: |
| 356 |
|
- remove 2nd set of calls at the top of initialise_varia.F (Adjoint part) |
| 357 |
|
related to Non_Lin Free-Surf or Bottom-Control (but not longer necessary). |
| 358 |
|
o verification/tutorial_tracer_adjsens/ |
| 359 |
|
- compile with ALLOW_CG2D_NSA |
| 360 |
|
- run with nonlinFreeSurf=3 |
| 361 |
|
- update reference output_adm.txt |
| 362 |
|
o tools/genmake2: |
| 363 |
|
- add "-DHAVE_NETCDF" to the DEFINES list when NetCDF is available ; |
| 364 |
|
refine usePROFILES switching off (only when NetCDF is not available). |
| 365 |
|
o verification (atmospheric set-up): |
| 366 |
|
- change eosType='IDEALGAS' to eosType='IDEALG' to match variable declaration |
| 367 |
|
o pkg/profiles : refine profiles and files handling. |
| 368 |
|
o model/src and pkg/ecco : introduce run-time switch usePROFILES. |
| 369 |
|
o model/src/the_main_loop.F, forward_step.F and pkg/exf/EXF_OPTIONS.h : |
| 370 |
|
- changes to use model/src/the_main_loop.F rather than the ecco version |
| 371 |
|
in ecco type adjoint runs. |
| 372 |
|
o model/src: |
| 373 |
|
- move computation of recip of grid distance & area from INI_MASKS_ETC to INI_GRID |
| 374 |
|
- move setting of h0Fac[C,W,S] from INI_LINEAR_PHISURF to INI_MASKS_ETC |
| 375 |
|
- update_masks_etc.F: rename SMOOTH*_R4,R8 function to the corresponding type (RS,RL) |
| 376 |
|
o model/src: |
| 377 |
|
- fix S/R UPDATE_CG2D for ALLOW_DEPTH_CONTROL case (was empty before); |
| 378 |
|
update output of test experiment bottom_ctrl_5x5. |
| 379 |
|
o pkg/generic_advdiff: |
| 380 |
|
- implement vertical bi-harmonic diffusion (for now, only for T & S and |
| 381 |
|
only explicitly). |
| 382 |
|
o verification/bottom_ctrl_5x5/ |
| 383 |
|
update reference output (results changed quite some time ago, |
| 384 |
|
not sure when; but new reference results look ok in terms of grdchk) |
| 385 |
|
o verification/1D_ocean_ice_column/ |
| 386 |
|
For verification purposes, the exact adjoint rather than an approx. |
| 387 |
|
should be used (e.g., to comare against F.D. or TLM). |
| 388 |
|
removed autodiff_inadmode*_ad.F routines in code_ad/ |
| 389 |
|
Update results/ accordingly. |
| 390 |
|
o pkg/cal: |
| 391 |
|
- change cal_getdate.F so that output date is function of input time only |
| 392 |
|
(less ambiguous, not mixed up with input iteration number). |
| 393 |
|
- fix calendarDumps for non-zero phase-lag (S/R CAL_TIME2DUMP). |
| 394 |
|
o pkg/diagnostics: |
| 395 |
|
- implement calendarDumps also for stats-diags |
| 396 |
|
- implement calendarDumps also for snap-shot output (negative freq) |
| 397 |
|
- for now, ignores calendarDumps if non-zero phase-lag. |
| 398 |
|
|
| 399 |
|
checkpoint63n (2012/06/04) |
| 400 |
|
o model/src: |
| 401 |
|
- recompute rSurfW & S to ensure consistent column thickness (needed for rStar) |
| 402 |
|
between Sum_k(hFac*drF) and rSurf-rLow, for now only when using pkg/shelfice; |
| 403 |
|
- use h0FacC, W & S to compute hFac_surfC, W & S in calc_surf_dr.F (more |
| 404 |
|
similar to calc_r_star.F); update results of experiment internal_wave and |
| 405 |
|
tutorial_global_oce_in_p (affected @ truncation level). |
| 406 |
|
o pkg/shelfice: |
| 407 |
|
- add parameterisation of advective-diffusive flux into the ice shelf |
| 408 |
|
according to Holland and Jenkins (1999), eq.22-33, as suggested by Jan De Ryd |
| 409 |
|
- simply eps3/3a computation |
| 410 |
|
o eesupp: |
| 411 |
|
- add new CPP option (DISCONNECTED_TILES) to disconnect tiles (no exchange |
| 412 |
|
between tiles, just fill-in edges assuming locally periodic subdomain) and |
| 413 |
|
add corresponding exchange subroutines (EXCH0); |
| 414 |
|
- new CG-solver version (_EX0) for disconnected-tiles special case. |
| 415 |
|
o model/src (CG-solvers): |
| 416 |
|
- modify Single-Reduction version (CG2D_SR) in case cg2dMaxIters is reached |
| 417 |
|
(similar final residual output and max-Iter number as with standard version); |
| 418 |
|
- allows to store and use lowest-residual solution (standard & _SR versions); |
| 419 |
|
- cleaning up (remove old, commented out code, LETS_MAKE_JAM ...). |
| 420 |
|
- by default, also compile CG2D_SR version (define ALLOW_SRCG in CPP_OPTIONS.h) |
| 421 |
|
o eesup/src: |
| 422 |
|
- global_vec_sum.F cleaning: add a BARRIER ; update comments ; remove ndef |
| 423 |
|
ALWAYS_USE_MPI (no longer needed after fixing usingMPI settings - March 2012) |
| 424 |
|
o pkg/salt_plume added to lab_sea/input_ad.noseaicedyn |
| 425 |
|
|
| 426 |
|
checkpoint63m (2012/04/27) |
| 427 |
|
o pkg/shelfice: |
| 428 |
|
- account for partical cells in surface level below ice-shelf for hydrostatic |
| 429 |
|
pressure calculion (set uniformFreeSurfLev to F); update isomip experiments. |
| 430 |
|
o ad test experiments: |
| 431 |
|
- update and rename to "ad_optfile.local" local AD-optfile specific to a |
| 432 |
|
particular adjoint verification experiment (in verification/*/code_ad/) |
| 433 |
|
o pkg/seaice & pkg/exf: |
| 434 |
|
- Part two of ECCO-CLIVAR code merges. |
| 435 |
|
Completes merge of seaice area relax. code. |
| 436 |
|
Requires unified CPP option EXF_ALLOW_SEAICE_RELAX |
| 437 |
|
to be set in EXF_OPTIONS.h since this pkg controls its I/O |
| 438 |
|
(and avoids second CPP option in SEAICE_OPTIONS.h) |
| 439 |
|
- change default value of costIceStart & costIceEnd to a valid date. |
| 440 |
|
- add snowPrecip to HSNOW in seaice_growth |
| 441 |
|
o pkg/ptracers: |
| 442 |
|
- move ptracers summary and setup check into separated S/R ; add a check for |
| 443 |
|
valid PTRACERS_Iter0 value (error+stop or just warning). |
| 444 |
|
o pkg/cal (+ pkg/ecco): |
| 445 |
|
- avoid calling a S/R with 2 times the same arg (FWD: unless both are only used |
| 446 |
|
as input; and for AD, unless both are not differentiable (e.g., k index)). |
| 447 |
|
o model/src: |
| 448 |
|
- define internal flag "uniformFreeSurfLev" (=T when free-surface level-index |
| 449 |
|
is everywhere k=1) to speed-up some calculations (by pass kSurf reference); |
| 450 |
|
temporary set to usingZCoords value (to be changed to F when using shelfice). |
| 451 |
|
- hydrostatic pressure: implement partial cell near the surface also for |
| 452 |
|
Z-coordinates (for the case where uniformFreeSurfLev=F); replace test on |
| 453 |
|
usingZCoords with test on uniformFreeSurfLev where needed (in NonHyd code). |
| 454 |
|
o pkg/ctrl/ctrl_init.F: |
| 455 |
|
- add a formal parameter to s/r ctrl_init_rec to print |
| 456 |
|
xx_$(ctrl_valiable)_file for easier debugging |
| 457 |
|
- move call of cal_* routines into the if-block where they are really |
| 458 |
|
needed so that xx_$(ctrl_variable)_period can be 0 or -12 without having |
| 459 |
|
to specify a startdate |
| 460 |
|
o pkg/seaice: |
| 461 |
|
- update comments in SEAICE_OPTIONS.h (+ similar update of lab_sea/code and |
| 462 |
|
global_ocean.cs32x15/code versions). |
| 463 |
|
- turn on seaice Multi-category (in data.seaice, SEAICE_multDim=7) for |
| 464 |
|
verification experiment global_ocean.cs32x15.seaice and update results. |
| 465 |
|
o pkg/exf: |
| 466 |
|
- new S/R EXF_GETFIELD_START to return starting-time of a forcing field |
| 467 |
|
(according to useExfYearlyFields). |
| 468 |
|
o pkg/cal: |
| 469 |
|
- move calendar setting & summary (previously in cal_readparams.F) into new |
| 470 |
|
S/R CAL_INIT_FIXED to be called later in the initialisation process, once |
| 471 |
|
length of the simulation is set (after SET_PARMS call); This is an attempt |
| 472 |
|
to fix Pb when, e.g.: pkg/longstep, modifies the length of the run. |
| 473 |
|
- For safety: keep a record of pkg/cal parameter settings status |
| 474 |
|
to allow to STOP if one pkg/cal S/R is called too early. |
| 475 |
|
- add a stop in all basic pkg/cal subroutines if called before setting is done. |
| 476 |
|
- modify cal_getdate.F: stop when nothing was returned; |
| 477 |
|
if called with myIter=-1, simply retrun modelBaseDate (1,2 = startDate_1,_2); |
| 478 |
|
this allows to remove "cal.h" from seaice_read_parms.F & init_mnc_vars.F. |
| 479 |
|
|
| 480 |
|
checkpoint63l (2012/04/04) |
| 481 |
|
o pkg/bbl: |
| 482 |
|
- fix call to FIND_RHO_2D in bbl_calc_rho.F (case multi-tiles per proc); |
| 483 |
|
update global_with_exf output.yearly.txt file. |
| 484 |
|
o pkg/cal: |
| 485 |
|
- start to clean-up cal_set.F ; fix for the case where baseTime is non-zero. |
| 486 |
|
- remove 2 unsused and very short (1 line only) functions ; |
| 487 |
|
- remove unsused subroutines from AD-Diff list. |
| 488 |
|
o model/src: |
| 489 |
|
- start to implement clean wait to stop in initialization S/R (config_check.F, |
| 490 |
|
packages_check.F, ini_parms.F and pkg/obcs/obcs_check.F + pkg/mypackage) |
| 491 |
|
o folowing PRINT_MESSAGE changes: |
| 492 |
|
- remove reminder multiple print to the same file (STDOUT) when multi-threads. |
| 493 |
|
o pkg/bbl: |
| 494 |
|
- modification so it can work with z* |
| 495 |
|
- replace deltaT with dTtracerLev(k) for asynchronous time stepping |
| 496 |
|
- global_with_exf/input.yearly verification; changes output.yearly.txt |
| 497 |
|
o eesupp/src: |
| 498 |
|
- PRINT_MESSAGE: use OpenMP critical instruction (similar to PRINT_ERROR); |
| 499 |
|
writing to unit 0 (if unit=errorMessageUnit) extended to the case myThid=1 |
| 500 |
|
- CHECK_THREADS: specific code if using OpenMP threading: improve checking of |
| 501 |
|
# of threads across processors (e.g., different env OMP_NUM_THREADS setting) |
| 502 |
|
- improve special preliminary call to EESET_PARMS (if eedata is read 2 times). |
| 503 |
|
- when using MPI, always set usingMPI=T ; set default accordingly and stop |
| 504 |
|
if not consistent. |
| 505 |
|
- ALL_PROC_DIE: add a BARRIER after MPI_FINALIZE call ; skip this when thread |
| 506 |
|
argument is zero (in case called within single-thread section). |
| 507 |
|
o pkg/fizhi: |
| 508 |
|
clean-up turbulence cold-start switch: decided in fizhi_init_vars.F, stored |
| 509 |
|
in common bloc (fizhi_coms.h) and then passed as argument up to S/R TURBIO. |
| 510 |
|
o pkg/exch2: |
| 511 |
|
- add argument "updateCorners" to S/R exch2_get_uv_bounds (enable to |
| 512 |
|
activate EXCH_IGNORE_CORNERS in vector EXCH S/R). |
| 513 |
|
- switch 1rst EXCH call to IGNORE_CORNERS (instead of UPDATE_CORNERS) |
| 514 |
|
but keep the 2nd with UPDATE_CORNERS: this prevents overwriting good data |
| 515 |
|
with bad ones coming from a not-yet-updated halo. |
| 516 |
|
This should fix the problem when sNx <> sNy and GCD(sNx,sNy) < 2*Olx. |
| 517 |
|
o pkg/frazil: |
| 518 |
|
move FrzForcT diag from frazil_diagnostics_state.F to frazil_calc_rhs.F |
| 519 |
|
http://mitgcm.org/pipermail/mitgcm-devel/2012-March/005384.html |
| 520 |
|
o model/src/config_check.F: add a check for OLx>sNx/OLy>sNy |
| 521 |
|
o verification/lab_sea |
| 522 |
|
* #undef SEAICE_EXCLUDE_FOR_EXACT_AD_TESTING |
| 523 |
|
* #define ALLOW_SST_CONTROL & ALLOW_SST_CONTROL |
| 524 |
|
* nTimeSteps 3 -> 4 |
| 525 |
|
* changed reference output accordingly |
| 526 |
|
o pkg/ctrl: |
| 527 |
|
* fix ctrl_get_gen for xx_sst, xx_sss |
| 528 |
|
o pkg/fizhi: |
| 529 |
|
- fix 2 calls to S/R GRD2MSC (argument igrd) in update_earth_exports.F |
| 530 |
|
update results of exp. fizhi-cs-32x32x40 (fail with only 6 digits for cg2d) |
| 531 |
|
- by default, use new version of S/R GETPWHERE (#define TRY_NEW_GETPWHERE); |
| 532 |
|
fix few out-of-bounds indices in referencing vars (initialise igrd to 1); |
| 533 |
|
clean-up fizhi_turb.F (remove unused & uninitialised qdiaglsm array). |
| 534 |
|
o pkg/gmredi: |
| 535 |
|
- add diagnostic of Redi (main diagnonal) vertical diffusive flux of Temp |
| 536 |
|
o pkg/fizhi: |
| 537 |
|
- fix few initialisation and multi-tile Pb: aqualev & gridalt now pass the |
| 538 |
|
(old)aces open64 tuv test. |
| 539 |
|
o pkg/icefront: fix bug and reorder loop in icefront_init_varia.F |
| 540 |
|
o pkg/mom_vecinv: |
| 541 |
|
- fix 1 loop range in mom_vecinv.F to be like all the other loops which |
| 542 |
|
set/update gu,gv/Diss (fix Pb of uninitialised fVerUkp in admom_vecinv |
| 543 |
|
with non-lin free-surf). |
| 544 |
|
o model/src: move k-loop outside of i/j-loops when (re-)computing R_low |
| 545 |
|
in ini/update_masks_etc.F in order to avoid -O3 optimization problems with |
| 546 |
|
some compilers (gfortran 4.6.0 on MacOS) |
| 547 |
|
o momentum: |
| 548 |
|
- separate fVer?(:,:,kUp) & fVer?(:,:,kDown) in argument list of MOM_FLUXFORM |
| 549 |
|
& MOM_VECINV subroutines (to help TAF). This fixes Pb of uninitialised |
| 550 |
|
fVer?(kUp) in admom_fluxform with non-lin free-surf or bottom control. |
| 551 |
|
|
| 552 |
|
checkpoint63k (2012/03/17) |
| 553 |
|
o model/src: |
| 554 |
|
- selectAddFluid (ALLOW_ADDFLUID): add missing term in horizontal momentum |
| 555 |
|
flux-form (mom_fluxform.F) and vertical momentum (calc_gw.F) equations. |
| 556 |
|
- SOLVE_DIAGONAL: switch default to use adjoinable & vectorized version |
| 557 |
|
(#undef SOLVE_DIAGONAL_LOWMEMORY); update output of exp. front_relax.bvp. |
| 558 |
|
o model/src: solve_*diagonal.F |
| 559 |
|
- replace division with multiplication by inverse (similar to original version) |
| 560 |
|
update output of AD exp. global_ocean.90x40x15. |
| 561 |
|
- fix pkg/gmredi/gmredi_calc_psi_bvp.F & pkg/ggl90/ggl90_calc.F |
| 562 |
|
to enable the use of the vectorized & differentiable solve_tridiagonal.F |
| 563 |
|
- test exp. global_ocean.90x40x15/input_ad: switch to advection scheme 20 for |
| 564 |
|
vertical advection (to get solve_tridiagonal tested in AD mode). |
| 565 |
|
o pkg/seaice: ECCO-CLIVAR merge: |
| 566 |
|
1. seaice AREA relaxation |
| 567 |
|
* CPP option: SEAICE_ALLOW_AREA_RELAXATION |
| 568 |
|
* runtime flag: SEAICE_tauAreaObsRelax |
| 569 |
|
2. modified global mean cost imbalance code: CPP option |
| 570 |
|
ALLOW_SSH_GLOBMEAN_COST_CONTRIBUTION |
| 571 |
|
o another iteration on solve_*diagonal.F |
| 572 |
|
Make choices independent of AUTODIFF |
| 573 |
|
Default will be the adjoinable vectorized version. |
| 574 |
|
Alternative choices are: |
| 575 |
|
* non-adjoinable original version with low memory footprint: |
| 576 |
|
#define SOLVE_DIAGONAL_LOWMEMORY |
| 577 |
|
* still adjoinable but does not vectorize (k as inner-most loop): |
| 578 |
|
#define SOLVE_DIAGONAL_KINNER (keeping SOLVE_DIAGONAL_LOWMEMORY undef) |
| 579 |
|
o testreport |
| 580 |
|
- add option to use a home made wrapper around "make" |
| 581 |
|
- remove MITGCM_IEEE env. variable (no longer active) |
| 582 |
|
o pkg/shelfice |
| 583 |
|
- improve computation of friction velocity for velocity dependent |
| 584 |
|
exchange coefficients: include velocities into SHELFICEBoundaryLayer |
| 585 |
|
block to average velocities vertically |
| 586 |
|
- add TAF store directives for SHI_ALLOW_GAMMAFRICT |
| 587 |
|
o Modify diagonal solvers, now three versions (upon request): |
| 588 |
|
#ifndef SOLVE_DIAGONAL_AUTODIFF |
| 589 |
|
1. default version (dating back a while ago) |
| 590 |
|
#else /* def SOLVE_DIAGONAL_AUTODIFF */ |
| 591 |
|
# ifdef SOLVE_DIAGONAL_KINNER |
| 592 |
|
2. adjoinable version with k-innermost loop (G. Forget) |
| 593 |
|
# else |
| 594 |
|
3. adjoinable vectorized version of 2. |
| 595 |
|
# endif |
| 596 |
|
#endif /* SOLVE_DIAGONAL_AUTODIFF */ |
| 597 |
|
All flags are currently set within routines, but could/should be moved to CPP-options. |
| 598 |
|
Array bounds for 1. are as before, for 2.,3. cover the entire domain |
| 599 |
|
(could be changed for 1. too). |
| 600 |
|
o verification/lab_sea (fwd): |
| 601 |
|
- update the 4 fwd lab_sea test experiment: no longer SEAICE_GROWTH_LEGACY, |
| 602 |
|
use 7 categories (except hb87), new solve4temp params (except salt_plume), |
| 603 |
|
constant salinity (=4) or variable salinity (salt_plume) done through |
| 604 |
|
SItracer. |
| 605 |
|
o pkg/seaice: |
| 606 |
|
- rename parameters: SIsalFrac to SEAICE_saltFrac & SIsal0 to SEAICE_salt0; |
| 607 |
|
- remove old parameters ( SEAICE_freeze, SEAICEturbFluxFormula, SEAICE_gamma_t, |
| 608 |
|
SEAICE_gamma_t_frz, SEAICE_availHeatTaper & SEAICE_availHeatFracFrz) |
| 609 |
|
from SEAICE_PARMS.h ; still need to fix seaice_cost_sst.F (broken when |
| 610 |
|
SEAICE_freeze got retired) but already better than previously with |
| 611 |
|
SEAICE_freeze=123456.7 deg.C; |
| 612 |
|
- move setting of facOpenGrow/Melt from SEAICE_INIT_FIXED to SEAICE_READPARMS |
| 613 |
|
(safer multi-threaded setting); always set SEAICEuseEVP; |
| 614 |
|
- setting of ocean-ice turb. flux coeff: moved from SEAICE_CHECK to |
| 615 |
|
SEAICE_READPARMS & SEAICE_INIT_FIXED (this fixes wrong summary report); |
| 616 |
|
stop if multiple specifications for the same coeff; make sure default |
| 617 |
|
SEAICE_mcPheePiston is compatible with drF(1) & deltaT. |
| 618 |
|
o pkg/obcs: |
| 619 |
|
- allow to switch to upwind 1rst order advection scheme for the advective |
| 620 |
|
flux computation at the open-boundary; disabled for adjoint built. |
| 621 |
|
o pkg/ptracers: |
| 622 |
|
- rename PTRACERS_RESTART.h -> PTRACERS_START.h |
| 623 |
|
- rename ptracers_fields_blocking_exchanges.F -> ptracers_fields_blocking_exch.F |
| 624 |
|
- add parameter to decide when to start a tracer; use internal flag to |
| 625 |
|
know which tracer to step-forward; switch this internal flag at the top |
| 626 |
|
of forward_step.F (new S/R PTRACERS_SWITCH_ONOFF); for now, not enabled |
| 627 |
|
for adjoint built. |
| 628 |
|
o pkg/seaice: |
| 629 |
|
- add SEAICE_SIZE.h anywhere in the code where SEAICE.h is included ; |
| 630 |
|
- move MULTDIM declaration and setting from SEAICE.h to SEAICE_SIZE.h ; |
| 631 |
|
- keep experiment offline_exf_seaice truly single category (set MULTDIM=1). |
| 632 |
|
o pkg/ecco: |
| 633 |
|
- fix writing of output file DiagnosCost_ClimSalt/Theta |
| 634 |
|
and DiagnosCost_DriftSalt/Theta (if multi-tiles per proc). |
| 635 |
|
o pkg/seaice: |
| 636 |
|
- add missing initialisation of COR_ICE array in dynsolver.F |
| 637 |
|
- merging SEAICE_MULTICATEGORY define and undef cases. Now the same code is |
| 638 |
|
compiled either way, and is tested in all exps (multidim was only tested |
| 639 |
|
in lab_sea ad before). |
| 640 |
|
- cosmetic change to prepare for future addition of ITD. no change to results. |
| 641 |
|
no defaults change. |
| 642 |
|
- added run time param SEAICE_multDim (7 for SEAICE_MULTICATEGORY, 1 otherwise). |
| 643 |
|
- by lack of adequate verification experiments, I further used custom versions |
| 644 |
|
of global_ocean.cs32x15 to test multicat or not, with pickup or not. All |
| 645 |
|
should be correct. |
| 646 |
|
- avoid recomputations in ad. |
| 647 |
|
o array initialisation in adjoint mode: |
| 648 |
|
1) pkg/mom_common/mom_calc_ke.F: |
| 649 |
|
add full array initialisation of KE inside mom_calc_ke.F (ALLOW_AUTODIFF_TAMC): |
| 650 |
|
The S/R which calls MOM_CALC_KE initialises KE but in some recomputation |
| 651 |
|
part this initialisation is gone, resulting in Floating Point Exception |
| 652 |
|
(caught by open64 compiler with debug option). |
| 653 |
|
2) model/src/convective_weights.F |
| 654 |
|
set output convectCount over the full array range (instead of doing |
| 655 |
|
an update): this fix a similar issue when called within recomputation (adjoint) |
| 656 |
|
which could by-pass the initialisation (normally done in the caller S/R). |
| 657 |
|
o pkg/seaice: |
| 658 |
|
- final version of turbulent flux params, as agreed at MIT ice meet: |
| 659 |
|
SEAICE_mcPheePiston (default is ~ 0.000875) |
| 660 |
|
SEAICE_mcPheeTaper (default is 0.) |
| 661 |
|
SEAICE_frazilFrac (defualt is 1.) |
| 662 |
|
SEAICE_mcPheeStepFunc (default is false) |
| 663 |
|
- changes results at trunction error level |
| 664 |
|
- ifdef SEAICE_EXTERNAL_FLUXES, disallow SEAICE_waterAlbedo |
| 665 |
|
o added pkg/frazil: |
| 666 |
|
- see frazil_description.tex for details |
| 667 |
|
- verification experiment is global_with_exf/input.yearly; changes output.yearly.txt |
| 668 |
|
o pkg/shelfice/shelfice_thermodynamics.F |
| 669 |
|
- fix exponents for velocity dependent exchange coefficients: change |
| 670 |
|
from 2/3(=0) to 0.66666666666666667 |
| 671 |
|
- move computation of velocity dependent exchange coefficients into |
| 672 |
|
separate loop, so that it applies to both thermodynamics |
| 673 |
|
- move some divisions by constants out of loops |
| 674 |
|
|
| 675 |
|
checkpoint63j (2012/02/16) |
| 676 |
|
o pkg/seaice: |
| 677 |
|
- overhaul of seaice_summary.F after feb 2012 clean-up of seaice_growth.F etc. |
| 678 |
|
- SItracer: added run time params, and allow coupling to ocn salinity. |
| 679 |
|
o pkg/seaice: remove (almost) all divisions by constants within loops |
| 680 |
|
in seaice_growth.F and seaice_solve4temp.F, this changes (only) the cg2d |
| 681 |
|
results of global_ocean.cs32x15.seaice from 16 to 11 digits (but only |
| 682 |
|
those, caused by 1/QI -> recip_QI) -> update output.seaice.txt, |
| 683 |
|
output_adm.seaice_dynmix.txt and also lab_sea: output_adm.evp.txt |
| 684 |
|
o pkg/seaice : |
| 685 |
|
- simplify McPhee tapering, and allow to use it with SEAICEturbFluxFormula |
| 686 |
|
= 1 or 2, using newly added run time parameter SEAICE_availHeatTaper. |
| 687 |
|
- fix d_AREAbyATM, d_AREAbyOCN, d_AREAbyICE diags I broke in r1.148. |
| 688 |
|
- add permanent SItflux diag that corresponds to TFLUX but includes |
| 689 |
|
ice+snow. Hence SItflux-TFLUX should match the ice+snow heat budget. |
| 690 |
|
- allow activation/testing of a fix for suspected missing term in |
| 691 |
|
ocn-ice heat budget (to be confirmed). To test this, you want to |
| 692 |
|
undef SEAICE_DISABLE_HEATCONSFIX, and then set the run time param |
| 693 |
|
SEAICEheatConsFix to .TRUE. that also allows the 'SIaaflux' diagnostic. |
| 694 |
|
- no change in results, fwd or ad. |
| 695 |
|
o global_ocean.cs32x15/results: 3 changes following from the pkg/seaice commit below: |
| 696 |
|
- switch to variable freezing point. |
| 697 |
|
Results change : cg2d -> 1 digits; ad -> 3-6-6 digits. |
| 698 |
|
- bug fix : the sublimation term that was missing is now activated. |
| 699 |
|
Results change : cg2d -> 2 digits; ad -> 8-6-6 digits. |
| 700 |
|
- bug fix : flooding sign term for SIsal0.NE.0. |
| 701 |
|
Result change : cg2d -> 8 digits; ad -> 11-12-9 digits. |
| 702 |
|
o pkg/seaice: |
| 703 |
|
- unified freezing point treatment : the old SEAICE_VARIABLE_FREEZING_POINT |
| 704 |
|
is now the default, and the old default constant freezing point is recovered |
| 705 |
|
with SEAICE_tempFrz0 = -1.96, |
| 706 |
|
SEAICE_dTempFrz_dS = 0., |
| 707 |
|
- retire SEAICE_freeze (= the old way of specifycing the constant freezing point) |
| 708 |
|
- remove ALLOW_SEAICE_FLOODING brackets; run time switch is already there. |
| 709 |
|
- bug fix (thanks to M. Losch) : the sublimation term that was missing is now |
| 710 |
|
activated. To reproduce old results with this bug define SEAICE_DISABLE_SUBLIM. |
| 711 |
|
- bug fix (silly me) : flooding sign term (seaice_growth.F r 1.149line 1211) |
| 712 |
|
was wrong for SIsal0.NE.0. Changes cs32x15 results (see above item for detail). |
| 713 |
|
- bug fix (thanks to O. Jahn) : area loss for melting got messed up in |
| 714 |
|
seaice_growth r149 for legacy branch (no results change). |
| 715 |
|
- introduce SEAICE_CAP_SUBLIM : caps sublimation heat flux in solve4temp (code |
| 716 |
|
from I. Fenty). |
| 717 |
|
- results did not change except for global_ocean.cs32x15, mostly due to the |
| 718 |
|
switch to variable freezing point (see above item for detail). |
| 719 |
|
o pkg/thsice: |
| 720 |
|
- fix and simplify interface to salt_plume pkg ; diagnose directly amount of |
| 721 |
|
seawater freezing in thsice_calc_thickn (not used). |
| 722 |
|
o pkg/seaice: turbulent flux term specification via run time params. |
| 723 |
|
- replace CPP switches with run time switches to facilitate testing |
| 724 |
|
- retired CPPs : MCPHEE_OCEAN_ICE_HEAT_FLUX, |
| 725 |
|
GRADIENT_MIXED_LAYER_TURBULENCE_FACTOR |
| 726 |
|
- added runtime parameter : SEAICEturbFluxFormula |
| 727 |
|
- move Mcphee constants to SEAICE_PARAMS.h |
| 728 |
|
- seaice_readparms.F : read turbulent flux term params (SEAICEturbFluxFormula; |
| 729 |
|
SEAICE_availHeatFrac/Frz, or SEAICE_gamma_t/_frz if apply); |
| 730 |
|
remove seaice_summary call (moved to seaice_init_fixed) |
| 731 |
|
- seaice_init_fixed.F : set SEAICE_availHeatFrac/Frz depending on |
| 732 |
|
SEAICEturbFluxFormula, and seaice_summary call (moved from seaice_readparms) |
| 733 |
|
- seaice_summary.F : add missing report of read turbulent flux term set up |
| 734 |
|
and SITRACER. |
| 735 |
|
- seaice_check.F : check range; inform and stop if retired CPPs are used |
| 736 |
|
(incl. SEAICE_AGE). |
| 737 |
|
- seaice_growth.F : unified use of SEAICE_availHeatFrac/Frz, without |
| 738 |
|
CPP switch, same computation as before for the different options. |
| 739 |
|
- change results at the truncation level. |
| 740 |
|
o pkg/seaice: |
| 741 |
|
- remove un-used parameters LAD & MAX_TICE ; add new run-time parameters: |
| 742 |
|
useMaykutSatVapPoly (default=F), postSolvTempIter (default=2) and |
| 743 |
|
SEAICE_wetAlbTemp (default=-1e-3). Remove option SEAICE_SOLVE4TEMP_LEGACY. |
| 744 |
|
- update verification/*/input*/data.seaice: |
| 745 |
|
to reproduce previous results with former #defined SEAICE_SOLVE4TEMP_LEGACY |
| 746 |
|
code, set useMaykutSatVapPoly=T & postSolvTempIter=0 ; |
| 747 |
|
and to reproduce previous results with former non-SOLVE4TEMP_LEGACY code, |
| 748 |
|
set SEAICE_wetAlbTemp=0 & SEAICE_snowThick=0. |
| 749 |
|
o pkg/seaice/seaice_solve4temp.F (SOLVE4TEMP_LEGACY): |
| 750 |
|
- remove LEGACY code for solving for tsurf (A1,A2,A3) but maintain the same |
| 751 |
|
algorithm (same choice: useMaykutPolySatVap=T, postSolvTempIter=0); |
| 752 |
|
difference in results only due to machine truncation. |
| 753 |
|
- remove MAX_TICE (tsurf is always =< TMELT anyway); keep MIN_TICE if using |
| 754 |
|
MaykutPolySatVap; keep MIN_LWDOWN. |
| 755 |
|
- adapt SEAICE_MODIFY_GROWTH_ADJ code (untested) to non-legacy formulation. |
| 756 |
|
- update output of verification exp. lab_sea (4 fwd: 11 cg2d digits except hb87, |
| 757 |
|
+ 3 AD: 6 digits except evp) and offline_exf_seaice.seaice (12 digits). |
| 758 |
|
o pkg/salt_plume: |
| 759 |
|
- isolate saltPlumeFlux array exch into a specific S/R called from |
| 760 |
|
do_oceanic_phys.F |
| 761 |
|
- add interface to pkg/salt_plume (filling of saltPlumeFlux) in pkg/thsice. |
| 762 |
|
o verification: |
| 763 |
|
- move untested set-up to MITgcm_contrib/verification_other. |
| 764 |
|
o pkg/seaice: |
| 765 |
|
- retire old way of SEAICE_AGE. Now replaced with ALLOW_SITRACER. |
| 766 |
|
o pkg/seaice: |
| 767 |
|
- seaice_solve4temp.F (keeping results unchanged): |
| 768 |
|
* add 2 future run-time params to select which saturation vapor pressure |
| 769 |
|
formula is used and how post-iteration fluxes are set. |
| 770 |
|
Temporary set these 2 params according to option SOLVE4TEMP_LEGACY; |
| 771 |
|
* New piece of code (not used) to update surf. atmos. heat flux F_ia using |
| 772 |
|
the linearized expression to stay consistent with tsurf finding. |
| 773 |
|
- seaice_freedrift.F avoid floating exception in sqrt(y) since y=sqrt(x*x)-x |
| 774 |
|
can be < 0. |
| 775 |
|
o pkg/seaice/seaice_solve4temp.F (SOLVE4TEMP_LEGACY undef): |
| 776 |
|
- remove from F_ia (and from it's derivative dFiDTs1) the contribution of |
| 777 |
|
conductive heat flux F_c , and add it explicitly when updating tsurf as |
| 778 |
|
solution of: Fc = Fia + d/dT(Fia - Fc) *delta.tsurf |
| 779 |
|
so that now F_ia has a consistent meaning through the entire routine. |
| 780 |
|
This causes differences at truncation level: update output of fwd exp. |
| 781 |
|
global_ocean.cs32x15.seaice. |
| 782 |
|
o pkg/seaice: |
| 783 |
|
- lots of editing/cleaning in seaice_solve4temp.F + better documentation |
| 784 |
|
of arrays, sign/direction of fluxes; no effect on results. |
| 785 |
|
- replace hard coded 273.16 (=previous value of celsius2K) by celsius2K |
| 786 |
|
in seaice_budget_ocean.F & seaice_solve4temp.F (SOLVE4TEMP_LEGACY part); |
| 787 |
|
note: this should have been done before changing celsius2K default. |
| 788 |
|
- to keep results unchanged, add back old value of celsius2K in verification |
| 789 |
|
exp. input*/data : lab_sea (AD), offline_exf_seaice (fwd+AD) & seaice_obcs |
| 790 |
|
- replace hard coded seawater freezing temp in seaice_solve4temp.F (part |
| 791 |
|
SOLVE4TEMP_LEGACY) of 271.20 (=previous value of celsius2K + default |
| 792 |
|
value of SEAICE_freeze) by celsius2K+SEAICE_freeze so that seawater |
| 793 |
|
freezing temp is consistent with value in seaice_growth.F ; this causes |
| 794 |
|
differences at machine truncation level: update lab_sea AD output (3/4). |
| 795 |
|
o pkg/seaice: |
| 796 |
|
- mask (in exf_init_fixed.F) atemp,aqh,evap,precip,runoff,lwflx,areamask |
| 797 |
|
and climsst/sss even when using seaice (to prevent calculation that might |
| 798 |
|
use SST on land); does not affect any verification exp. |
| 799 |
|
- apply mask to taux,tauy (in seaice_get_dynforcing.F); update output |
| 800 |
|
of lab_sea.fd & lab_sea.hb87 and global_ocean.cs32x15.seaice_dynmix. |
| 801 |
|
o pkg/seaice/seaice_growth.F: add mask to avoid non-zero ice thickness |
| 802 |
|
over land (non-legacy code) |
| 803 |
|
o vectorize salt_plume package: |
| 804 |
|
- add extra code to salt_plume_frac and salt_plume_tendency_apply_s |
| 805 |
|
- this code computes plumebk everywhere (also over land), thus requires |
| 806 |
|
code to avoid divisions by zero and catch the case of kLev=Nr. |
| 807 |
|
- efficiency of salt_plume_frac is still sub-optimal |
| 808 |
|
o tools (genmake2): |
| 809 |
|
- fix makefile NAME specification (option: -makefile NAME) by adding |
| 810 |
|
"-f $MAKEFILE" in few places where it was missing including makedepend |
| 811 |
|
command (contribution from Paul). |
| 812 |
|
|
| 813 |
|
checkpoint63i (2012/01/24) |
| 814 |
|
o model/src: |
| 815 |
|
- add factorized versions of some EOS code to find_rho.F, |
| 816 |
|
define USE_FACTORIZED_EOS to turn it on (default for TARGET_NEC_SX) |
| 817 |
|
o model/src: |
| 818 |
|
- add celsius2K to namelist PARM01; remove unused recip_rhoNil from PARAMS.h; |
| 819 |
|
- change default celsius2K from 273.16 to 273.15 |
| 820 |
|
- to keep results unchanged, add back old value of celsius2K in verification |
| 821 |
|
exp. input*/data : aim.5l_cs, 1D_ocean_ice_column, lab_sea (fwd), |
| 822 |
|
global_ocean.cs32x15(seaice) & cpl_aim+ocn(input_atm). |
| 823 |
|
o model/src: add a little initialisation trick to find_rho.F and find_alpha.F |
| 824 |
|
to make TAF generate vectorizable code |
| 825 |
|
o pkg/seaice/seaice_evp.F: add code (copied from CICE) that avoids underflows. |
| 826 |
|
Turn on by defining SEAICE_EVP_ELIMINATE_UNDERFLOWS. Most compilers |
| 827 |
|
have flags that do this more efficiently. |
| 828 |
|
o pkg/diagnostics/diagstats_calc.F |
| 829 |
|
- for TARGET_NEC_SX fix the treatment of the scaling factor tmpFac=scaleFact |
| 830 |
|
o pkg/obcs (with exch2): |
| 831 |
|
- derive OB indices in overlap region from OB-InsideMask (in obcs_init_fixed.F) |
| 832 |
|
and remove code (in obcs_readparms.F) which was based on EXCH of indices |
| 833 |
|
(through buffer) and did not account for exch2 (e.g., CS-grid) topology. |
| 834 |
|
o pkg/exf and seaice exp: |
| 835 |
|
- change offline_exf_seaice.seaicetd test exp. to use default DIFF1 value (=0) |
| 836 |
|
so that switching off area & heff advection flags does not affect results. |
| 837 |
|
- change default EXF_LWDOWN_WITH_EMISSIVITY to #define in EXF_OPTIONS.h |
| 838 |
|
- keep exp. seaice_obcs & 1D_ocean_ice_column(ad) unchanged (using #undef |
| 839 |
|
EXF_LWDOWN_WITH_EMISSIVITY) and, using pkg/exf default emissivities, update |
| 840 |
|
results of verification experiments: 1D_ocean_ice_column(fwd), |
| 841 |
|
global_ocean.cs32x15.seaice(fwd+ad) & seaice_dynmix(ad), lab_sea (all 8) |
| 842 |
|
and offline_exf_seaice (all 4). |
| 843 |
|
o pkg/seaice: |
| 844 |
|
- remove array YNEG, RIVER, TMIX (replaced by local var) from common block; |
| 845 |
|
remove also areaNm1 & hEffNm1 when SEAICE_GROWTH_LEGACY is undef. |
| 846 |
|
- fix bug: areaNm1 & hEffNm1 are needed in seaice_growth (SEAICE_GROWTH_LEGACY) |
| 847 |
|
but were not filled when resp. SEAICEadvArea=F & SEAICEadvHeff=F and |
| 848 |
|
not multi-dim advection; |
| 849 |
|
- fix seaice_growth.F to compile with both SEAICE_GROWTH_LEGACY and |
| 850 |
|
SHORTWAVE_HEATING #undef. |
| 851 |
|
o optim: make Makefile more robust (?) |
| 852 |
|
o pkg/obcs: |
| 853 |
|
* fix treatment of obcs_read_pickup for TAF |
| 854 |
|
o pkg/shelfice: |
| 855 |
|
* add missing initialisation of num_shifwflx |
| 856 |
|
* new adjoint verification exp. for isomip |
| 857 |
|
* change Prandtl and Schmidt numbers from generic formulations |
| 858 |
|
(in terms of model diffus. and visc.) to separate runtime |
| 859 |
|
parameters, with Holland and Jenkins (1999) default values. |
| 860 |
|
o pkg/exf (interpolation): |
| 861 |
|
- extended input field (2 rows) near the N & S poles: |
| 862 |
|
* fill in with the symetric value (when even Nb of data in longitude) |
| 863 |
|
* add average value at the poles only for scalar quantities ; for vector |
| 864 |
|
component interpolation, skip the averaging and keep duplicated values, |
| 865 |
|
unless the 2 components have same location (uvInterp=T) in which case |
| 866 |
|
interpolate the 2 components together using S/R EXF_INTERP_UV. |
| 867 |
|
- new S/R EXF_INTERP_UV to interpolate the 2 components of a vector field |
| 868 |
|
together: this allows to account properly for local orientation when |
| 869 |
|
averaging at N & S pole. |
| 870 |
|
- add CPP option EXF_USE_OLD_INTERP_POLE to recover old results regarding |
| 871 |
|
exf-interpolation near N & S poles. Undef by default. |
| 872 |
|
- add debug check/print on 1rst iter if debugLevel >=2. |
| 873 |
|
o pkg/exf: |
| 874 |
|
- change masking flags definition (no longer "parameter"), put them in |
| 875 |
|
common block and set them in S/R EXF_INIT_FIXED. |
| 876 |
|
- change recognized masking flag from s,u,v to c,w,s since the mask |
| 877 |
|
to use (maskC,W,S) depends on position on model grid and has little to |
| 878 |
|
do with scalar/vector difference; fix masking if using P-coordinates. |
| 879 |
|
- fix masking of wind-stress (was wrong if interpolated or read on A-grid) |
| 880 |
|
and update results of exp. global_with_exf (x2). |
| 881 |
|
- fix setting of zenith-angle table when SHORTWAVE_HEATING is undef. |
| 882 |
|
o pkg/obcs |
| 883 |
|
- S/R obcs_calc_stevens: fix a small bug (gammat -> gammas where appropriate) |
| 884 |
|
o pkg/exf (interpolation): |
| 885 |
|
- assume periodicity in X only if input field cover full longitude range. |
| 886 |
|
- improve search for lat. index (supposed to be faster, in ~log2(ny) steps, |
| 887 |
|
and should vectorise). |
| 888 |
|
- fix input lat of the 2 added row (in case we provide N.pole data). |
| 889 |
|
|
| 890 |
|
checkpoint63h (2011/12/30) |
| 891 |
|
o pkg/seaice: |
| 892 |
|
- change declaration of UVM from _RL to _RS in order to get B-grid |
| 893 |
|
uIce,vIce monitor when _RS = real*4 |
| 894 |
|
o pkg/offline: |
| 895 |
|
- new S/R OFFLINE_RESET_PARMS for resetting main-model params (switch off flags); |
| 896 |
|
- compute vertical velocity if not read from file (switch on exactConserv |
| 897 |
|
& staggerTimeStep); |
| 898 |
|
- fix output of ptracer monitor iter & time for default PTRACERS_monitorFreq. |
| 899 |
|
o verification: |
| 900 |
|
- update grid-files (in tutorial_held_suarez_cs/input and fizhi-cs-32x32x40/input) |
| 901 |
|
with grid-angle generated using "utils/matlab/cs_grid/cubeCalcAngle.m": |
| 902 |
|
fix Pb in yG definition @ 2 missing corners (previously yG was averaged); |
| 903 |
|
- update output of exp. global_ocean.cs32x15.viscA4 after changing |
| 904 |
|
grid-file angles (used here for coriolis-3d); |
| 905 |
|
- remove (or gzip) few files not needed to run fizhi-cs experiments. |
| 906 |
|
o pkg/exf (vector interpolation): |
| 907 |
|
- rotate vector component toward local grid-axis orientation using grid-angles |
| 908 |
|
(stored in GRID.h) instead of locally computed rotation angle; keep old code |
| 909 |
|
available with exf option EXF_USE_OLD_VEC_ROTATION (undef by default). |
| 910 |
|
o pkg/seaice: |
| 911 |
|
- allow to compile with exf options ALLOW_ATM_TEMP or ALLOW_DOWNWARD_RADIATION |
| 912 |
|
undefined. Fix and remove the stop when ALLOW_RUNOFF is undefined. |
| 913 |
|
o pkg/PKG_OPTIONS.h files: |
| 914 |
|
- move #include "CPP_OPTIONS.h" just after #include "PACKAGES_CONFIG.h", |
| 915 |
|
outside of the #ifdef ALLOW_THISPKG / #endif block. |
| 916 |
|
- update verification/*/code/PKG_OPTIONS.h files. |
| 917 |
|
o add a verification experiment (offline_exf_seaice/input.seaice) that test |
| 918 |
|
the B-grid LSOR solver |
| 919 |
|
o pkg/seaice: sort out emissivity and long wave radiation step 2: |
| 920 |
|
- add emissivity for snow and ice as new run-time parameters |
| 921 |
|
(unfortunately) that default to the corresponding exf values; this |
| 922 |
|
changes results in most verification experiments so that for backward |
| 923 |
|
compatibility these parameter have to be set explicitly to inconsistent |
| 924 |
|
values (ugly); |
| 925 |
|
- the exf-flag EXF_LWDOWN_WITH_EMISSIVITY then gets rid off the hard-wired |
| 926 |
|
emissivities of 0.97 associated with lwdownloc in S/R seaice_solve4temp |
| 927 |
|
until we agree how to handle this bug properly |
| 928 |
|
o pkg/exf (interpolation): |
| 929 |
|
- add a set of run-time param ({inputfield}_interpMethod), one for |
| 930 |
|
each interpolated input field, to select the interpolation method |
| 931 |
|
with a value of zero switching off the interpolation; |
| 932 |
|
- print interpolation parameters to standard output; |
| 933 |
|
- refine exf_check.F accordingly ; stop if USE_NO_INTERP_RUNOFF is defined |
| 934 |
|
(to use instead runoff_interpMethod=0 in data.exf). |
| 935 |
|
- simplify default setting of interpolation parameters (for all fields, |
| 936 |
|
set position to match model grid cell-center position, leading to trivial |
| 937 |
|
interpolation in case of regular Lat-Lon grid) and fix for the case of |
| 938 |
|
non-uniform delY. |
| 939 |
|
o pkg/exf/exf_radiation.F: add the factor ocean_emissivity to lwdown, for |
| 940 |
|
within #ifdef EXF_LWDOWN_WITH_EMISSIVITY/#endif. Unless ocean_emissivity=1. |
| 941 |
|
this factor is absolutely required, because the lw exitance (= out-going |
| 942 |
|
long wave radiation) is emissivity*stefanBoltzmann*T^4 + rho*lwdown, |
| 943 |
|
where the reflectivity rho = 1-emissivity for conservation reasons. Once |
| 944 |
|
this is established, we can remove the CPP-Flag (if we want to). |
| 945 |
|
o pkg/seaice: |
| 946 |
|
- qualify a stop statement in S/R seaice_check: EXF_READ_EVAP and |
| 947 |
|
ALLOW_BULKFORMULAE are only required if SEAICE_EXTERNAL_FLUXES is defined |
| 948 |
|
- sort out emissivity and long wave radiation, step 1: make SEAICE_emissivity |
| 949 |
|
what the name implies (and not emissivity*BoltzmannConstant), this means |
| 950 |
|
the old default value of 5.5e-8 now means |
| 951 |
|
5.5e-8/5.670e-8 = 0.97001763668430343479 as ocean_emissivity in |
| 952 |
|
S/R exf_readparms |
| 953 |
|
- FV-discretization of the B-grid LSOR solver, turn on with |
| 954 |
|
SEAICE_LSRBNEW. This new solver also involves a slight re-organization |
| 955 |
|
of the calling routine dynsolver, so that B and C-grid code look a |
| 956 |
|
little more alike. |
| 957 |
|
- fix index bugs in ostres.F (B-grid code) |
| 958 |
|
o pkg/seaice/seaice_ocean_budget.F: change (unused) code to make it look better |
| 959 |
|
o pkg/seaice/seaice_readparms.F: make ocean_emissivity*stefanBoltzmann the |
| 960 |
|
default for SEAICE_emissivity, if useEXF; requires setting a wrong |
| 961 |
|
parameter value in global_ocean.cs32x15/input.seaice/data.seaice for |
| 962 |
|
backward compatiblity |
| 963 |
|
o model/src: use sigmaR in calc_ivdc |
| 964 |
|
|
| 965 |
|
checkpoint63g (2011/12/16) |
| 966 |
|
o pkg/seaice/seaice_growth.F: |
| 967 |
|
fix heat conservation of the coupled ocean-seaice system. |
| 968 |
|
when SEAICE_HEAT_CONSERV_FIX is defined, the mismatch of |
| 969 |
|
solid vs liquid water treatment is resolved via QNET. |
| 970 |
|
o pkg/seaice/seaice_growth.F: |
| 971 |
|
add missing term for sublimation in saltFlux, which was breaking conservation. |
| 972 |
|
o pkg/seaice/seaice_model.F: |
| 973 |
|
Remove this patholigical case treatment in evolution branch, |
| 974 |
|
where it is un-necessary and breaks conservation. |
| 975 |
|
Update global_ocean.cs32x15/results accordingly for |
| 976 |
|
output.seaice.txt output_adm.seaice.txt and output_adm.seaice_dynmix.txt |
| 977 |
|
o model/src: |
| 978 |
|
useRealFreshWaterFlux: remove masking of EmPmP (using maskInC) in |
| 979 |
|
external_forcing_surf.F and apply mask directly in S/R SOLVE_FOR_PRESSURE. |
| 980 |
|
o pkg/obcs: |
| 981 |
|
modify Stevens boundary conditions so that S/R obcs_calc_stevens computes |
| 982 |
|
tendency for temperature and salinity that is then applied to update |
| 983 |
|
theta and salt in S/R obcs_apply_ts; this requires proper masking of all |
| 984 |
|
horizontal advection and diffusion contributions on the boundary (in |
| 985 |
|
S/Rs gad_advection, gad_calc_rhs, gmredi_rtransport, gmredi_calc_diff), and |
| 986 |
|
removing the mask in S/R timestep_tracer. The advantage of this approach is |
| 987 |
|
that the updated tracer values include contributions from the surface |
| 988 |
|
forcing and vertical diffusion. |
| 989 |
|
This modification changes exp4.stevens because of vertical diffusion. |
| 990 |
|
o model/src: |
| 991 |
|
change non-curvilinear grid setting to also work with exch2 + blank tiles: |
| 992 |
|
- move delX,delY to new header file (SET_GRID.h) and adjust length to 1rst |
| 993 |
|
face dimensions. |
| 994 |
|
- new S/R INI_LOCAL_GRID to define model tile-local grid (using same units |
| 995 |
|
as delX,delY) and called from S/R INI_CARTESIAN_GRID, INI_CYLINDER_GRID |
| 996 |
|
and INI_SPHERICAL_POLAR_GRID. |
| 997 |
|
change test-experiment global_ocean.90x40x15: |
| 998 |
|
- split the domain in many more tiles (9x4 tiles of 10x10 each); |
| 999 |
|
- switch to exch2 and remove 1 blank-tile (but only for MPI test). |
| 1000 |
|
- update results of both test (std & dwnslp). |
| 1001 |
|
o pkg/diagnostics/diagstat_calc.F: fix problem with non-initialised halos in |
| 1002 |
|
tmp-fields for TARGET_NEC_SX at the cost of extra loops and less vectorisation |
| 1003 |
|
o model/src: |
| 1004 |
|
- take sigma-coords related code out of update_etah.F into new S/R UPDATE_ETAWS |
| 1005 |
|
and fix initialisation when using sigma-coords. |
| 1006 |
|
o model/src: |
| 1007 |
|
- improve Gael's changes from Jan 14, 2011 "addition of an 'update' of hfacc etc. |
| 1008 |
|
to their current value at the beginning of forward_step.F": add a new run-time |
| 1009 |
|
switch "doResetHFactors" (default=false) and use it to do (or not) the resetting. |
| 1010 |
|
- set doResetHFactors=T in the following test exp: |
| 1011 |
|
global_ocean.90x40x15/input (rStar) & input.dwnslp (linFS), |
| 1012 |
|
global_ocean.cs32x15/input.seaice (rStar), hs94.cs-32x32x5/input.impIGW (linFS) |
| 1013 |
|
and tutorial_global_oce_in_p (nonLinFS in p) |
| 1014 |
|
o pkg/ecco |
| 1015 |
|
- added IES cost term |
| 1016 |
|
o verification/advect_xz |
| 1017 |
|
- scale down the depth by 1/100 (from 200 km to just 2.km) |
| 1018 |
|
- add secondary test using divergent flow, r-star and implicit vertical advection. |
| 1019 |
|
- add note (check_conserve_TS.txt) and matlab script to check global |
| 1020 |
|
conservation of T & S with NonLin-FreeSurf and Adams-Bashforth. |
| 1021 |
|
o pkg/rbcs: |
| 1022 |
|
- add missing relaxUFile & relaxVFile in namelist (in rbcs_readparms.F). |
| 1023 |
|
o pkg/obcs (Stevens BCs): |
| 1024 |
|
- fix various problems in the phase speed computations of the Stevens |
| 1025 |
|
boundary conditions: |
| 1026 |
|
- sign error for western and southern boundaries, |
| 1027 |
|
- time tendencies were underestimated by factor of dTracerLev and some just |
| 1028 |
|
zero (for multidim advection schemes) -> now estimate time tendencies |
| 1029 |
|
from difference between previous (n) time level and the one before (n-1); |
| 1030 |
|
this requires storing a field per tracer per boundary, also in the pickup |
| 1031 |
|
files. |
| 1032 |
|
- replace deltaTmom in cflMer/Zon computation with dTracerLev |
| 1033 |
|
- set default phase velocity to cflMer/Zon |
| 1034 |
|
- update verification experiment exp4.stevens |
| 1035 |
|
o pkg/generic_advdiff (implicit_r): |
| 1036 |
|
- fix implicit vertical advection conservation with AB and/or NonLin-FreeSurf |
| 1037 |
|
(remove Tr*d/dz(w) in gad_implicit_r.F and add it in gad_calc_rhs.F); |
| 1038 |
|
- fix implicit vertical diffusion or advection with NonLin FreeSurf |
| 1039 |
|
(was using current recip_hFacC instead of future recip_hFacC): set future |
| 1040 |
|
recip_hFac in thermodynamics.F and pass & use it in all vertical implicit |
| 1041 |
|
solver S/R ; |
| 1042 |
|
- fill diagnostics of vertical advective fluxes (computed from tendency) |
| 1043 |
|
when using implicit vert. advection; |
| 1044 |
|
- update output of exp that uses NonLin-FreeSurf & implicit vertical diffusion: |
| 1045 |
|
global_ocean.90x40x15 (fwd+AD), global_ocean.cs32x15 (4 fwd +4 AD), |
| 1046 |
|
tutorial_global_oce_in_p (fwd), tutorial_tracer_adjsens (AD) |
| 1047 |
|
and cpl_aim+ocn (used in ocean); to be fixed: bottom_ctrl_5x5. |
| 1048 |
|
o pkg/diagnostics: |
| 1049 |
|
- add diags for T & S tendency which goes through Adams-Bashforth. |
| 1050 |
|
|
| 1051 |
|
checkpoint63f (2011/11/30) |
| 1052 |
|
o pkg/obcs/obcs_calc_stevens.F: |
| 1053 |
|
- join some loops to reduce number of if-statements |
| 1054 |
|
- fix a bug in code for northern and southern boundaries. |
| 1055 |
|
o pkg/seaice/seaice_init_varia.F |
| 1056 |
|
- undo resetting UVM=0 that has slipped in unnoticed in r1.49 and |
| 1057 |
|
caused zero ice-velocities for the B-grid code |
| 1058 |
|
o pkg/seaice/seaice_init_fixed.F |
| 1059 |
|
- fix a bug in the initialisation of the second metric terms |
| 1060 |
|
k2AtC and k2AtZ: k2AtC was set to the value of k2AtZ and k2AtZ was 0. |
| 1061 |
|
- this affects all seaice configurations with useCurviLinearGrid = .true. |
| 1062 |
|
and SEAICEuseDynamics = .true. (default) and SEAICEuseMetricTerms |
| 1063 |
|
= .true. (default) |
| 1064 |
|
- the only verification experiment affected is global_ocean.cs32x15: |
| 1065 |
|
icedyn, seaice, and seaice_dynmix (AD) |
| 1066 |
|
o pkg/thsice/thsice_diagnostics_state.F |
| 1067 |
|
- only for TARGET_NEC_SX: initialise all of tmpFld, so that overlap does |
| 1068 |
|
not contain garbage that can lead to surprises in diagstats_calc.F |
| 1069 |
|
o model/src: |
| 1070 |
|
- move bi,bj loops inside INTEGR_CONTINUITY ; |
| 1071 |
|
- call INTEGR_CONTINUITY directly from forward_step.F |
| 1072 |
|
(previously called from momentum_correction_step.F) |
| 1073 |
|
- call UPDATE_ETAH from integr_continuity.F and update initialise_varia.F, |
| 1074 |
|
forward_step.F and pkg/ecco/the_main_loop.F |
| 1075 |
|
o model/src/do_oceanic_phys.F: |
| 1076 |
|
- reset addMass to zero only if useICEFRONT ; |
| 1077 |
|
- separated k-loop for rhoInSitu calculation ; |
| 1078 |
|
- add 2 diagnostics for wVel*delta^k(Rho) at constant pressure (= WdRHO_P) |
| 1079 |
|
and at constant T,S (= WdRHOdP); |
| 1080 |
|
NOTE: change diagnostics WRHOMASS : this is now simply the Vertical transport |
| 1081 |
|
of In-Situ Density (Anomaly) (=wVel*rho^bar-k) (previously was Vertical |
| 1082 |
|
transport of locally-referenced potential density); |
| 1083 |
|
to recover old diag: (WRHOMASS)_old = (WRHOMASS)_new -1/2 * (WdRHOdP) |
| 1084 |
|
o gluemnc: |
| 1085 |
|
- replace ncrcat with ncks when adding variables to *.glob.nc |
| 1086 |
|
- change the way variables without record dimensions are handled so that |
| 1087 |
|
files that have variables with and without record variables can be glued: |
| 1088 |
|
add a record dimension with a new unique name to be able to differentiate |
| 1089 |
|
between variable with and without a-prior record dimensions. Remove |
| 1090 |
|
this record dimension before adding the variable to the *.glob.nc file. |
| 1091 |
|
This procedure automatically works also for grid.*.nc files that do not |
| 1092 |
|
have any record dimension. |
| 1093 |
|
o verification/cpl_aim+ocn: |
| 1094 |
|
- generate new reference output on baudelaire (gfortran, mpich2) (previously |
| 1095 |
|
from old aces-grid cluster, ao + ifort) and update 'run_cpl_test' script. |
| 1096 |
|
o pkg/obcs/obcs_calc_stevens.F: |
| 1097 |
|
- add code for northern and southern boundaries. |
| 1098 |
|
- reorder some loops to save some operations and so that drFBar can be scalar |
| 1099 |
|
o model/src: |
| 1100 |
|
- re-arrange extrapolation calculation in Adams-Bashforth S/R (affects machine |
| 1101 |
|
truncation) and add diagnostics for Adams-Bashforth tendency increment. |
| 1102 |
|
- update test-exp. AD output: |
| 1103 |
|
1D_ocean_ice_column, bottom_ctrl_5x5, lab_sea.noseaicedyn, obcs_ctrl; |
| 1104 |
|
and FWD output: exp4 (x3), fizhi-cs-32x32x40, fizhi-cs-aqualev20, |
| 1105 |
|
front_relax (x3), global_ocean.90x40x15 (x2), global_ocean.cs32x15 (x5), |
| 1106 |
|
global_ocean_ebm, ideal_2D_oce, internal_wave, isomip (x3), lab_sea (x4), |
| 1107 |
|
matrix_example, MLAdjust (x5), natl_box (x2), tutorial_advection_in_gyre, |
| 1108 |
|
tutorial_global_oce_in_p, tutorial_global_oce_latlon and |
| 1109 |
|
tutorial_plume_on_slope. |
| 1110 |
|
o pkg/diagnostics/diagstats_calc.F: add special code for TARGET_NEC_SX |
| 1111 |
|
that vectorises. This code uses 3 extra 2D fields, F90 array assignments |
| 1112 |
|
and F90 intrinsic routines SUM, MINVAL, MAXVAL, but since TARGET_NEC_SX |
| 1113 |
|
does not even come with an F77 compiler anymore, this should not be |
| 1114 |
|
a limitation |
| 1115 |
|
|
| 1116 |
|
checkpoint63e (2011/11/06) |
| 1117 |
|
o pkg/seaice adjoint of LSR: |
| 1118 |
|
- substantially rework STORE dirs. |
| 1119 |
|
- add new flag SEAICE_LSR_ADJOINT_ITER and parameter SOLV_MAX_FIXED |
| 1120 |
|
this flag will do SOLV_MAX_ITERS but only store SOLV_MAX_FIXED fields |
| 1121 |
|
and only perform SOLV_MAX_FIXED adjoint iters. |
| 1122 |
|
If flag SEAICE_LSR_ADJOINT_ITER is undef-ined, the store in the iteration |
| 1123 |
|
always goes to same record, meaning that the last forward field will be |
| 1124 |
|
the one used in the adjoint - an approximation. |
| 1125 |
|
- now produces non-zero LSR-related gradients, and acceptable gradients for |
| 1126 |
|
An's zonal (x-z) test case (to be checked in) |
| 1127 |
|
- lab_sea: produces acceptable gradients for lab_sea with forward/backward LSR |
| 1128 |
|
(but these are screwed up by the thermodynamics in the first place); |
| 1129 |
|
reduce from 4 to 3 timesteps to avoid t.d. issue as temporary solution; |
| 1130 |
|
output_adm.txt also reflects changes to seaice_lsr |
| 1131 |
|
- global_ocean.cs32x15: |
| 1132 |
|
* the "default" input_ad.seaice has been re-defined to always exclude dynamics |
| 1133 |
|
(output_adm.seaice.txt was updated) |
| 1134 |
|
* new input_ad.seaice_dynmix tests a forward LSR / backward free-drift |
| 1135 |
|
(output_adm.seaice_dynmix.txt) |
| 1136 |
|
o test-exp. exp4+stevens: |
| 1137 |
|
- switch off pTracers and nonHydrostatic (restart problems with NH); update output |
| 1138 |
|
o pkg/seaice (lsr solver): |
| 1139 |
|
- put coriolis term into FORCEX,Y (as in seaice_evp.F; previously in rhsU,V only); |
| 1140 |
|
update output (change truncation) of FWD test exp lab_sea & global_ocean.cs32x15 |
| 1141 |
|
- fix SEAICEuseFREEDRIFT for new OBCS version (ndef OBCS_UVICE_OLD) |
| 1142 |
|
- new parameter "LSR_mixIniGuess" allows (if=2,4) to mix free-drift solution |
| 1143 |
|
into initial guess for linearised solver; use weight ~ (1/err)^2,4. |
| 1144 |
|
- compute (with new S/R) and print initial (+ free-driff sol. if available) |
| 1145 |
|
and final residual of linearised system. |
| 1146 |
|
o pkg/pp81: add diagnostics for vertical eddy viscosity and diffusivity |
| 1147 |
|
(requires new subroutines pp81_init_fixed.F/pp81_diagnostics_init.F) |
| 1148 |
|
o pkg/obcs: Stevens BCs (still only for east and west) |
| 1149 |
|
- fix Stevens BCs by using intermediate velocities of previous |
| 1150 |
|
time step as "baroclinic" contribution rather velocity of one grid cell |
| 1151 |
|
inside the boundaries; more consistent but makes the entire treatment |
| 1152 |
|
more complicated, requires new pickup files and two new S/Rs: |
| 1153 |
|
obcs_copy_uv_n and obcs_save_uv_n |
| 1154 |
|
keep old version with flag OBCS_STEVENS_USE_INTERIOR_VELOCITY |
| 1155 |
|
(undefined by default) |
| 1156 |
|
- silently move computation of some grid factors into separate loops to |
| 1157 |
|
save some recomputations |
| 1158 |
|
- add a hack for non-linear free surface |
| 1159 |
|
- add a verification experiment to exp4 that tests this code |
| 1160 |
|
o make gluemnc/xplodemnc work with (moron) non-GNU shell tools |
| 1161 |
|
|
| 1162 |
|
checkpoint63d (2011/10/25) |
| 1163 |
|
o pkg/obcs + seaice-dynamics: |
| 1164 |
|
- new implementation calls OBCS_APPLY_UVICE (in seaice_dynsolver.F) before |
| 1165 |
|
solving for uIce,vIce and prevent SEAICE_LSR or SEAICE_EVP to modify OB |
| 1166 |
|
values (consistent with an implicit method). |
| 1167 |
|
- add option OBCS_UVICE_OLD (in OBCS_OPTIONS.h, defined by default) to |
| 1168 |
|
return to older implementation (same results as before); note that none of |
| 1169 |
|
the "experimental" OBCS_SEAICE options are implemented in the new version. |
| 1170 |
|
- switch test exp. seaice_obcs to use new version (update output). |
| 1171 |
|
o pkg/seaice: |
| 1172 |
|
- add few stop (B-grid & CS-exch, B-grid & obcs, AUTODIFF & SEAICE_maskRHS) |
| 1173 |
|
- add parenthesis in Water-Ice drag calculations (evp & lsr); |
| 1174 |
|
- in seaice_lsr.F, simplify and fix AA3 setting in case sNx or sNy=1; |
| 1175 |
|
and fix case SEAICE_waterDrag=0 (but unstable if no or little seaice) |
| 1176 |
|
- update output of test exp lab_sea that uses evp or lsr (fwd+ad) |
| 1177 |
|
and global_ocean.cs32x15 (icedyn & seaice fwd+ad). |
| 1178 |
|
o pkg/layers: |
| 1179 |
|
- Include potential density as new coordinate (Thanks to David Munday) |
| 1180 |
|
o pkg/dic: |
| 1181 |
|
- Remove subroutine CALC_PCO2_APPROX_CO3 from carbon_chem.F and add |
| 1182 |
|
carbonate computation/output to CALC_PCO2_APPROX |
| 1183 |
|
o pkg/generic_advdiff |
| 1184 |
|
- make TAF generate code that vectorizes: |
| 1185 |
|
gad_dst3fl_adv_*.F, gad_u3c4_impl_r.F |
| 1186 |
|
o model/src: |
| 1187 |
|
- improve vectorization also for adjoint code of |
| 1188 |
|
solve_tridiagonal and solve_pentadiagonal. As this requires up to |
| 1189 |
|
5 new 3D-fields, this code is only turned on for the combination of |
| 1190 |
|
ALLOW_AUTODIFF_TAMC and TARGET_NEC_SX. Also remove error message |
| 1191 |
|
for singular matrix check (only for TARGET_NEC_SX). |
| 1192 |
|
- remove hard-wiring of CPP-flag ALLOW_SOLVERS_KLOOPINSIDE to |
| 1193 |
|
ALLOW_AUTODIFF. This flag now needs to be set outside of |
| 1194 |
|
solve_tri/pentadiagonal.F |
| 1195 |
|
o pkg/exf: |
| 1196 |
|
- fix previous modif in exf_readparms.F (revision 1.66: treatment of |
| 1197 |
|
optional namelist EXF_NML_OBCS). |
| 1198 |
|
- add many debug-print (to trace down where pkg/cal error msg comes from) |
| 1199 |
|
o pkg/mom_vecinv/mom_vi_hdissip.F |
| 1200 |
|
- rearrange loops so that vectorization is improved and in particular |
| 1201 |
|
the adjoint routine vectorizes. The code is a little longer now, but I |
| 1202 |
|
think easier to read. |
| 1203 |
|
Note: CPP-flag MOM_VI_ORIGINAL_VISCA4 is not tested. |
| 1204 |
|
o pkg/seaice: |
| 1205 |
|
- allow to specify initial seaice velocity from files (uIceFile,vIceFile) |
| 1206 |
|
o pkg/exch2: |
| 1207 |
|
- alternative code (without tile x tile matrix) for cumulated (global) sum; |
| 1208 |
|
used as default; keep using matrix version in adjustment.cs-32x32x1. |
| 1209 |
|
o tools/build_options: |
| 1210 |
|
- after merging some MPI and/or multi-threads specific optfiles so that |
| 1211 |
|
the modified default optfile support both built (using genmake2 -mpi |
| 1212 |
|
for the MPI built, using genmake2 -omp for OpenMP built), |
| 1213 |
|
move now obsolete optfiles to tools/build_options/unsupported/ |
| 1214 |
|
(add copy of email sent to support list in doc/optfiles_changes.txt) |
| 1215 |
|
|
| 1216 |
|
checkpoint63c (2011/09/28) |
| 1217 |
|
o pkg/ebm: |
| 1218 |
|
- fix ebm_zonalmean.F for several procs in Ydir (nPy > 1); update output of |
| 1219 |
|
test exp. global_ocean_ebm (affects machine truncation, 12 digits for cg2d) |
| 1220 |
|
o pkg/seaice (SItracer pickup) |
| 1221 |
|
* Simple code for pickup of SItracer fields (#define ALLOW_SITRACER) |
| 1222 |
|
To be done: |
| 1223 |
|
- thorough testing of missing fields in S/R seaice_read_pickup |
| 1224 |
|
- thorough matching between generic tracers and specific fields used |
| 1225 |
|
(e.g. volume-weighted age, area-weighted age, salinity, ...) |
| 1226 |
|
o pkg/grdchk: |
| 1227 |
|
- all procs calculate and print finite-difference gradient (for testreport) |
| 1228 |
|
- change description output used by testreport; |
| 1229 |
|
- AD testing: also compare and report Finite-difference gradient; |
| 1230 |
|
=> update all output_adm*.txt |
| 1231 |
|
o test exp. global_ocean_ebm: |
| 1232 |
|
- switch back useRealFreshWaterFlux (no problem anymore with run-off); |
| 1233 |
|
and turn on exactConserve; update (fwd) output. |
| 1234 |
|
o pkg/seaice (LSR): |
| 1235 |
|
- print to STDOUT number of solver iterations at SEAICE_monFreq interval. |
| 1236 |
|
- modify solver to deal cleanly with u/v switch at face edges (CS-grid): |
| 1237 |
|
* identical results with non-CS topology (but save up to half of EXCH calls) |
| 1238 |
|
* no much changes in cs-32 verification tests since no ice at face edges |
| 1239 |
|
(differences due to convergence criteria, since nb of iterations changed) |
| 1240 |
|
* expect some differences on lat-lon-cap grid. |
| 1241 |
|
update FWD results of global_ocean.cs32x15 test exp. (icedyn & seaice) |
| 1242 |
|
o pkg/exch2: |
| 1243 |
|
- add new CPP option to disable Cumul-Sum matrix (to save storage when using |
| 1244 |
|
many tiles); until alternative coding is added, just add a stop. |
| 1245 |
|
o pkg/kpp: |
| 1246 |
|
- move ptracer/salt correction to KPP diffusivity from calc_3d_diffusivity.F |
| 1247 |
|
to S/R KPP_CALC_DIFF_PTR (+ add tracer-index to argument list). |
| 1248 |
|
o optfiles: |
| 1249 |
|
- merged few MPI specific (*+mpi*) and multi-threads specific (*+mth) optfiles |
| 1250 |
|
into corresponding standard one, which now support different build: |
| 1251 |
|
(when using "genmake2 -mpi" for MPI built, "genmake2 -omp" for OpenMP built) |
| 1252 |
|
- add "darwin_amd64_gfortran" optfile (initially from Constantinos, tested by Gael) |
| 1253 |
|
o pkg/obcs (seaice): |
| 1254 |
|
- add missing EXCH for uIce,vIce after applying OBCS; |
| 1255 |
|
- remove S/R EXF_SWAPFFIELDS_X/Y , disable S/R EXF_SET_OBCS_X/Y by adding |
| 1256 |
|
argument = number of levels to S/R EXF_SWAPFFIELDS_XZ/YZ and S/R |
| 1257 |
|
EXF_SET_OBCS_XZ/YZ ; update obcs_exf_load.F; |
| 1258 |
|
- fix EXF_SET_OBCS_XZ/YZ for multi-threaded (assuming loaded arrays are shared) |
| 1259 |
|
- verification exp. seaice_obcs: split domain in 2 tiles ; add SIZE.h_mpi |
| 1260 |
|
and eedata.mth for MPI and mutlti-threaded testing. |
| 1261 |
|
o pkg/flt: |
| 1262 |
|
- make FLT EXCH working with multi-threads (and MPI + multi-threads); |
| 1263 |
|
- use buffer for floats IO (trajectories & profiles); size set in FLT_SIZE.h; |
| 1264 |
|
- fix pickup and trajectories + profiles output with multi-threads; |
| 1265 |
|
- add parameter to select group of fields to write along trajectory and |
| 1266 |
|
profile (default unchanged); |
| 1267 |
|
- only call PORT_RAND* function when flt_noise <> 0; |
| 1268 |
|
- add multi-threaded test for verification exp. flt_example. |
| 1269 |
|
o eesupp/src: |
| 1270 |
|
- fix S/R ALL_PROC_DIE for MPI+multi-threads. |
| 1271 |
|
- USE_OMP_THREADING: check that number of threads (from eedata) matches |
| 1272 |
|
actual number (from env.var. setting). |
| 1273 |
|
o pkg/ebm: |
| 1274 |
|
- fix multi-threading in ebm_area_t.F ; clean-up EBM.h header file. |
| 1275 |
|
- add multi-threaded test for verification exp. global_ocean_ebm |
| 1276 |
|
|
| 1277 |
|
checkpoint63b (2011/08/29) |
| 1278 |
|
o pkg/ebm: |
| 1279 |
|
- fix sign of runoff; |
| 1280 |
|
- update directly forcing fields (Qnet,EmPmR) to use standard |
| 1281 |
|
external_forcing_surf.F (remove ebm_forcing_surf.F); |
| 1282 |
|
- avoid loading forcing fields a 2nd time (1rst time from ini_forcing.F); |
| 1283 |
|
- set either tauThetaClimRelax (in data) or tauThetaZonRelax (in data.ebm) |
| 1284 |
|
and apply relaxation towards corresponding SST field (2-D map or ZonAver). |
| 1285 |
|
- update global_ocean_ebm parameters file data and update output (fwd+ad). |
| 1286 |
|
o verification/matrix_example: |
| 1287 |
|
- split domain in 8 tiles (2x4); add SIZE.h_mpi and eedata.mth for |
| 1288 |
|
MPI and mutlti-threaded testing. |
| 1289 |
|
o pkg/matrix: |
| 1290 |
|
- change name of S/R MATRIX_INIT to S/R MATRIX_INIT_VARIA ; |
| 1291 |
|
- fix for multi-tiles (add bi,bj indices to exp/impMatrixCounter); |
| 1292 |
|
- fix for multi-threaded (remove DATA statement); |
| 1293 |
|
- use standard DIFFERENT_MULTIPLE function; |
| 1294 |
|
- add MATRIX_OPTIONS.h |
| 1295 |
|
o verification/matrix_example: |
| 1296 |
|
To allow to split the domain in several tiles, change the domain size |
| 1297 |
|
from 31x31 to 32x32 by adding 1 row (on Southern side) + 1 column (on |
| 1298 |
|
Western side) of land points. |
| 1299 |
|
o (*_)write_pickup: |
| 1300 |
|
- fix few bugs in mypackage_read/write_pickup.F |
| 1301 |
|
- in MDS_WR_METAFILES calls, replace "myTime" by a local copy into |
| 1302 |
|
array of length 1 (was caught by compiler syntax-check). |
| 1303 |
|
o pkg/diagnostics (velocity potential solver): |
| 1304 |
|
- fix initialisation of main diagonal matrix (for overlap values) |
| 1305 |
|
o tools/build_options: |
| 1306 |
|
- towards single optfile for MPI and non-MPI built: merge few +mpi optfile |
| 1307 |
|
into previously non MPI one (linux_am64 pgf77,open64,pathf90,ifort11) |
| 1308 |
|
- linux_amd64_gfortran: merge +mpi_generic version into standard optfile: |
| 1309 |
|
* move IEEE flags to DEVEL section; take IEEE flags from +mpi_generic ; |
| 1310 |
|
* FAST case: for now, keep -O3 with ini_masks_etc.F in NOOPTFILES list; |
| 1311 |
|
and use "-devel" testreport option for standard testing on baudelaire. |
| 1312 |
|
o pkg/grdchk: |
| 1313 |
|
- Assume that all gradient-check position are within the same tile |
| 1314 |
|
corresponding to global indices (iGloTile,jGloTile) and that |
| 1315 |
|
positions (i,j indices or nbeg) are local to this tile; |
| 1316 |
|
- set tile local-indices and proc. number according to tile global-indices; |
| 1317 |
|
- remove grdchkwhichproc from namelist (retired param); |
| 1318 |
|
- send adjoint gradient to all procs so that all procs print it to STDOUT; |
| 1319 |
|
- switch on MPI test for global_ocean.90x40x15 AD test experiment. |
| 1320 |
|
o adding package bbl (Bottom Boundary Layer) |
| 1321 |
|
description in MITgcm/pkg/bbl/bbl_description.tex |
| 1322 |
|
example/test experiment in MITgcm_contrib/bbl |
| 1323 |
|
o adjoint verifications |
| 1324 |
|
- changed 5 experiments to multi-tiles |
| 1325 |
|
* global_ocean.90x40x15 |
| 1326 |
|
* obcs_ctrl |
| 1327 |
|
* offline_exf_seaice |
| 1328 |
|
* tutorial_dic_adjoffline |
| 1329 |
|
* tutorial_tracer_adjsens |
| 1330 |
|
changed reference output for those that crossed tile boundaries |
| 1331 |
|
(changed choices of gradient check positions) |
| 1332 |
|
- with same tiling, switch on MPI test for the last 4 (add SIZE.h_mpi) |
| 1333 |
|
- temporary bug fix in grdchk_loc.F for obcs |
| 1334 |
|
o genmake2 & testreport |
| 1335 |
|
- add target "obj" (just for OBJFILES files) to Makefile; |
| 1336 |
|
- add option to specify FEXTRAFLAGS on command line. |
| 1337 |
|
o pkg/mypackage: |
| 1338 |
|
- remove bi,bj arguments from argument list of S/R MYPACKAGE_CALC_RHS |
| 1339 |
|
|
| 1340 |
|
checkpoint63a (2011/08/02) |
| 1341 |
o pkg/diagnostics (velocity potential): |
o pkg/diagnostics (velocity potential): |
| 1342 |
- fix velocity-potential calculation when using OBCS: balance net inflow |
- fix velocity-potential calculation when using OBCS: balance net inflow |
| 1343 |
at each level (new S/R OBCS_DIAG_BALANCE). |
at each level (new S/R OBCS_DIAG_BALANCE). |
| 1388 |
Add Sub-Meso Eddies parameterisation from Baylor (for now, disabled with TAF), |
Add Sub-Meso Eddies parameterisation from Baylor (for now, disabled with TAF), |
| 1389 |
and test it in verification exp. "front_relax.bvp" (update output.bvp.txt) |
and test it in verification exp. "front_relax.bvp" (update output.bvp.txt) |
| 1390 |
|
|
| 1391 |
checkpoint63 (2011/07/11) |
checkpoint63 (2011/07/11) |
| 1392 |
o pkg/exch2: |
o pkg/exch2: |
| 1393 |
- rename+move: nTiles in W2_EXCH2_PARAMS.h --> exch2_nTiles in W2_EXCH2_TOPOLOGY.h |
- rename+move: nTiles in W2_EXCH2_PARAMS.h --> exch2_nTiles in W2_EXCH2_TOPOLOGY.h |
| 1394 |
- exch2 version of Cumul-Sum routine: S/R W2_SET_MAP_CUMSUM set up the |
- exch2 version of Cumul-Sum routine: S/R W2_SET_MAP_CUMSUM set up the |
Parent Directory
| 
Revision Log
| 
Revision Graph
| 
Patch