| 4 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
| 5 |
============================== |
============================== |
| 6 |
|
|
| 7 |
o eesup and pkg/mdsio: ALLOW_WHIO |
o pkg/seaice: |
| 8 |
|
- introduce SEAICEareaFormula run time parameter, which selects the recipe for |
| 9 |
|
computing the ice cover tendency from effective thickness tendencies. Include Ian Fenty's formula. |
| 10 |
|
- remove 'undef SEAICE_OCN_MELT_ACT_ON_AREA' which may now be achieved using SEAICEareaFormula. |
| 11 |
|
- simplify notations for the various d_HEFF/d_AREA terms. |
| 12 |
|
o pkg/gmredi: |
| 13 |
|
- add Boundary-Value-Problem bolus transport (Ferrari etal, OM 2010) code |
| 14 |
|
from Christopher Wolfe (UCSD); add partial cell factor in vertical derivative; |
| 15 |
|
add a verification test-run (front_relax.bvp) to test GM_useBVP. |
| 16 |
|
o pkg/seaice: |
| 17 |
|
- for SEAICE_DO_OPEN_WATER_GROWTH, replace Gael's implementation with Ian's. |
| 18 |
|
- the new SEAICE_DO_OPEN_WATER_MELT allows open-water air-sea heat fluxes to also melt ice. |
| 19 |
|
o pkg/seaice: |
| 20 |
|
- McPhee formula for Ice-Ocean fluxes (provided by Ian Fenty). |
| 21 |
|
- can be used by #define MCPHEE_OCEAN_ICE_HEAT_FLUX in SEAICE_OPTIONS.h |
| 22 |
|
o pkg/cheapaml: |
| 23 |
|
- from Nico Wienders: New version of this package. |
| 24 |
|
o pkg/flt: |
| 25 |
|
- add CPP options to prevent floats to re-enter the opposite side of |
| 26 |
|
a periodic domain (default: keep periodicity) |
| 27 |
|
o pkg/gmredi: |
| 28 |
|
- apply exchange to GM_PsiX,GM_PsiY to get valid values near CS-corners |
| 29 |
|
(needed for multi-dim advection). |
| 30 |
|
- switch on GM Advective form in global_ocean.cs32x15 and update output.txt |
| 31 |
|
o model/src: |
| 32 |
|
- fix wrong re-setting of addMass array (ALLOW_ADDFLUID) for atmospheric run. |
| 33 |
|
o tutorial_tracer_adjsens |
| 34 |
|
- extend to NLFS and replace maskC by hFacC |
| 35 |
|
(change reference output accordingly) |
| 36 |
|
|
| 37 |
Option to write/read 2D fields to files including tiles halos. |
checkpoint62r (2011/01/24) |
| 38 |
The main purpose is for adjoint related "tape I/O". |
o testreport: |
| 39 |
The secondary purpose is for debugging phases. |
- add option -deloutp (-do) to remove output file after successful run |
| 40 |
|
- add option -mfile (-mf) for MPI machile-file and for each test-exp, |
| 41 |
|
extract exact list of procs from this file ; adjust COMMAND by |
| 42 |
|
replacing TR_MFILE & TR_NPROC with local machile-file & Nb of MPI procs. |
| 43 |
|
o pkg/ecco: replace old ineffective way of penalizing deviations from |
| 44 |
|
first guess (of obcs) with code by Matt Mazloff |
| 45 |
|
o pkg/ctrl: add code that allows the use of ALLOW_OBCS?_CONTROL without |
| 46 |
|
exf/cal |
| 47 |
|
- ctrl_init: fix initialisation (computation of diffrec) for obcs-ctrl |
| 48 |
|
and handle case of obcs?period == 0 |
| 49 |
|
Not affected by these changes: There is still an inconsistency in |
| 50 |
|
the computation of start/endrec for the obcs variables for ALLOW_CAL |
| 51 |
|
defined (see comment by Menemenlis, 7-Mar-2003) |
| 52 |
|
- make S/R ctrl_get_gen_rec return something meaningful even if |
| 53 |
|
ALLOW_CAL is not defined |
| 54 |
|
- ctrl_getobcs?: simplify code so that it does not need to use |
| 55 |
|
S/R exf_swapfields |
| 56 |
|
o testreport: |
| 57 |
|
allow to test MPI with different number of processors for each experiment: |
| 58 |
|
create file "SIZE.h.mpi" in each build dir. |
| 59 |
|
Note: update of "-command" argument not yet implemented. |
| 60 |
|
o model/src, pkg/autodiff, pkg/ecco, pkg/seaice: |
| 61 |
|
alleviate the need for additional 3D tapes when using |
| 62 |
|
the non-inear free surface in the adjoint. This is done |
| 63 |
|
by adding an 'update' of hfacc etc. to their current |
| 64 |
|
value at the beginning of forward_step.F. |
| 65 |
|
o pkg/seaice |
| 66 |
|
seaice_lsr: move stress-divergence computations out of outer pseudo- |
| 67 |
|
timestepping loop |
| 68 |
|
o pkg/obcs/obcs_prescribe_read.F: move 'CALL OBCS_EXTERNAL_FIELDS_LOAD' |
| 69 |
|
up, so that the calls of the ctrl_* subroutines are really at the end |
| 70 |
|
of the routine. |
| 71 |
|
o pkg/gmredi: |
| 72 |
|
allow to specify a 2-D horizontal map and a 1-D vertical profil |
| 73 |
|
as scaling factor for Isopycnal diffusivity (Redi) and GM diffusivity. |
| 74 |
|
o pkg/seaice: |
| 75 |
|
move lsr iterations from seaice_dynsolver to seaice_lsr. |
| 76 |
|
o pkg/seaice: |
| 77 |
|
reduce memory footprint of seaice mutldim advection tapes, |
| 78 |
|
by using a 2D tape independent of the generic_advdiff ones, |
| 79 |
|
and setting GAD_AREA=1 etc. rather than GAD_AREA=101 etc. |
| 80 |
|
|
| 81 |
|
checkpoint62q (2010/12/29) |
| 82 |
|
o model/src: |
| 83 |
|
- check horizontal grid-spacing for unset or zero/negative values. |
| 84 |
|
- change AD test hs94.1x64x5 to be truly a 2-D set-up. |
| 85 |
|
o pkg/flt: |
| 86 |
|
- use (recent) mdsio S/R MDS_READVEC_LOC instead of local FLT_MDSREADVECTOR |
| 87 |
|
- precision of input and output files set according to readBinaryPrec |
| 88 |
|
and writeBinaryPrec (previously always 64.b), pickup remains at 64 bits. |
| 89 |
|
- initial float position from tiled file was not converted to index space |
| 90 |
|
(looks like a bug): fixed to make it similar to initial pos. global file. |
| 91 |
|
o global_ocean.cs32x15: |
| 92 |
|
change SIZE.h of AD set-up to get same tiling as with MPI. |
| 93 |
|
Update output_adm, output_adm.thsice (pass@14) & output_adm.seaice (fail@11) |
| 94 |
|
o model/src: |
| 95 |
|
add flag to control printing of domain fields (bathymetry, hFac ...) in |
| 96 |
|
ini_mask_etc (call to plot_fieds_xy*). |
| 97 |
|
o pkg/thsice: |
| 98 |
|
if useEXF, some forcing fields are not valid in overlap: fixed by computing |
| 99 |
|
over the interior only + apply EXCH at the end (like with SEAICE advection). |
| 100 |
|
o pkg/mdsio: |
| 101 |
|
- add 2 arguments to S/R MDS_PASS_R4/8toRL/S for overlap size of buffer |
| 102 |
|
to copy to/from, so that it can be used to replace MDS_PASS_WH_R8toRL/RS. |
| 103 |
|
- change barrier call for safe multi-threads access to 3-D shared buffer. |
| 104 |
|
- MDS_READ/WRITEVEC_LOC use buffers from MDSIO_BUFF_3D.h (instead of local ones) |
| 105 |
|
- replace call to MDS_PASS_WH_R4/8toRL with call to MDS_PASS_R4/8toRL and |
| 106 |
|
remove the 4 S/R MDS_PASS_WH_R4/8toRL/S. |
| 107 |
|
o pkg/flt: |
| 108 |
|
- add fourth-order Runge-Kutta integration |
| 109 |
|
- add basic exch2 support (only 1 facet supported so far!) |
| 110 |
|
o pkg/mdsio+rw: |
| 111 |
|
new S/R to read local (or tile independent) vector (read counterpart of |
| 112 |
|
mds_writevec_loc, and front-end write_glvec_rl/rs). |
| 113 |
|
o global_ocean.cs32x15: additional input_ad.seaice experiment |
| 114 |
|
that uses pkg/seaice rather than pkg/thsice. |
| 115 |
|
(using multidim advection for seaice) |
| 116 |
|
o towards more F95 syntax for array specification in namelist: |
| 117 |
|
- change "data.diagnostics" (data.ptracers was changed before checkpoint62p) |
| 118 |
|
of all verification experiments. |
| 119 |
|
e.g.: line: fields(1,3)='ETAN ','ETANSQ ', |
| 120 |
|
was changed to: fields(1:2,3)='ETAN ','ETANSQ ', |
| 121 |
|
- finish adding "-DNML_EXTENDED_F77" to DEFINES list in pgf77 and g77 optfiles |
| 122 |
|
o global_ocean.cs32x15: additional input.seaice experiment |
| 123 |
|
that uses pkg/seaice rather than pkg/thsice. |
| 124 |
|
|
| 125 |
|
checkpoint62p (2010/12/17) |
| 126 |
|
o pkg/thsice: rename (ad tape index) iicekey as ticekey, and move it |
| 127 |
|
to THSICE_SIZE.h to avoid conflict with pkg/seaice. |
| 128 |
|
o pkg/seaice: add seaice_diffusion.f seaice_map_thsice.f to adjoint. |
| 129 |
|
allow for use of multim. advection in seaice adjoint. |
| 130 |
|
o pkg/seaice: add diagnostics for snow precipitation and sublimation flux |
| 131 |
|
- first attempt to close budgets by taking into account the freshwater |
| 132 |
|
flux implied by the sublimation of snow and ice. So far only the |
| 133 |
|
volume budget is closed, turn on with soon-to-be-retired-again CPP-Flag |
| 134 |
|
SEAICE_ADD_SUBLIMATION_TO_FWBUDGET |
| 135 |
|
o eesupp/src: |
| 136 |
|
new S/R with new option (NML_EXTENDED_F77) to change array specification |
| 137 |
|
from F95 standard to commonly accepted extended F77 form. |
| 138 |
|
This would allow to use F95 syntax in namelist (needed with gfortran 4.4.5) |
| 139 |
|
and to revert back to the original syntax (with -DNML_EXTENDED_F77) |
| 140 |
|
for less tolerant f77 compiler (e.g., pgf77, g77). |
| 141 |
|
o globalFiles: |
| 142 |
|
commented out in parameter file "data" since it's not safe with MPI |
| 143 |
|
(e.g., wrong on aces); keep it if useSingleCpuIO=T (but not doing much |
| 144 |
|
in this case); no longer commented out in restart test (tst_2+2). |
| 145 |
|
o add adTapeDir to specify a special directory where adjoint tapes are |
| 146 |
|
written/read (adread_adwrite.F). adTapeDir is specified in data/PARM05 |
| 147 |
|
and cannot be used together with mdsioLocalDir. |
| 148 |
|
o nesting pkgs: |
| 149 |
|
- packages_init_variables : put call to OBCS_INIT_VARIABLES in last position |
| 150 |
|
- update Gianmaria's Nesting pkgs (from Contrib/nesting_sannino) |
| 151 |
|
- add parameter files + utils dir + output from short run on beagle. |
| 152 |
|
o pkg/seaice: |
| 153 |
|
fix seaice_growth.F store directives; update results of exp. lab_sea.ad |
| 154 |
|
o pkg/seaice: explicitly store tsurfloc in seaice_solve4temp.F to avoid |
| 155 |
|
unnecessary copying in adjoint code, requires new parameter NMAX_ITER |
| 156 |
|
|
| 157 |
|
checkpoint62o (2010/11/25) |
| 158 |
|
o pkg/exf: |
| 159 |
|
- change CPP option EXF_VERBOSE to run-time logical param exf_verbose. |
| 160 |
|
o pkg/seaice: |
| 161 |
|
- seaice_ocean_stress: split a loop for better vectorization of the |
| 162 |
|
adjoint code |
| 163 |
|
- seaice_growth.F: fix bug in conservation of freshwater connected to |
| 164 |
|
capping of heff, put contribution into d_HEFFbyNEG |
| 165 |
|
- replace irritating parameters SEAICE_latentWater/Ice and SEAICE_sensHeat |
| 166 |
|
by something more sensible (parameters that are what their name implies) |
| 167 |
|
- change some defaults, so that by default exf-parameters are used for |
| 168 |
|
things like rhoAir, cpAir,latent/sensible heat parameters, in order |
| 169 |
|
to reproduce the old results, many data.seaice now contain funny values |
| 170 |
|
for cpAir and latent heat of sublimation |
| 171 |
|
o packages_readparms.F: move exf_readparms up in line so that |
| 172 |
|
exf-parameters are visible for other packages (e.g. seaice) |
| 173 |
|
o debugMode & debugLevel=2: |
| 174 |
|
- differentiate between debugMode & debugLevel=2: |
| 175 |
|
debugMode: controls printing of debug messages (sequence of S/R calls). |
| 176 |
|
debugLevel: controls printing of algorithm intermediate results and stats. |
| 177 |
|
- remove "if debugLevel=2 => debugMode" but keep "debugMode => debugLevel=2"; |
| 178 |
|
=> allows to print debug-stats without all sequence of call messages. |
| 179 |
|
- switch control of debug print from debugLevel to debugMode in many S/R. |
| 180 |
|
o pkg/obcs (obcs-sponge): |
| 181 |
|
fix bug that was introduced on Oct 25 (in dynamics.F v1.151): put back the |
| 182 |
|
call to OBCS_PRESCRIBE_EXCHANGES (need to exch gU,gV when using obcs-sponge) |
| 183 |
|
o pkg/generic_advdiff: |
| 184 |
|
- record overlap minimum size for each advection scheme; |
| 185 |
|
- refine criteria on overlap minimum size; |
| 186 |
|
- check and stop (in gad_check) if overlap size is too small. |
| 187 |
|
o pkg/obcs: |
| 188 |
|
- with ptracers: remove multiple calls to obcs_apply_ptracer (only done |
| 189 |
|
once, in ptracers_implicit.F) |
| 190 |
|
- with seaice: remove test on empty file name in obcs_apply_seaice (always |
| 191 |
|
apply OB values when tile has OB) |
| 192 |
|
o model/src, coriolis setting: |
| 193 |
|
- integer parameter "selectCoriMap" (=0,1,2) replaces 3 logical |
| 194 |
|
useConstantF, useBetaPlaneF, useSphereF ; can now read coriolis from |
| 195 |
|
files (selectCoriMap=3) with any grid. |
| 196 |
|
- add parameter "fPrime" for constant second Coriolis coeff (default=0). |
| 197 |
|
o pkg/seaice/seaice_growth.F: |
| 198 |
|
a few cosmetic changes that reduce the number of lines of code |
| 199 |
|
o pkg/obcs: add boundary conditions following Stevens (1990) |
| 200 |
|
- the code follows old MOM/NAOSIM code |
| 201 |
|
- so far it is only implemented for the eastern and western boundaries |
| 202 |
|
and NOT for pTracers |
| 203 |
|
o pkg/rbcs: |
| 204 |
|
- replace rbcsInIter by rbcsForcingOffset, the time in seconds at the |
| 205 |
|
beginning of the first rbcs forcing period. |
| 206 |
|
- add option rbcsSingleTimeFiles that makes rbcs read forcing data like the |
| 207 |
|
offline package, i.e., one file per time, labeled by iteration number; |
| 208 |
|
deltaTrbcs sets the timestep with which the data was generated; |
| 209 |
|
rbcsIter0 is the forcing iteration corresponding to rbcsForcingOffset. |
| 210 |
|
- rbcsForcingCycle=0 makes forcing non-cyclic. |
| 211 |
|
- make arithmetic safe for negative times and odd forcing periods. |
| 212 |
|
|
| 213 |
|
checkpoint62n (2010/11/09) |
| 214 |
|
o pkg/seaice: |
| 215 |
|
fix masking in free-drift S/R ; update results of exp. lab_sea.fd |
| 216 |
|
o pkg/obcs: |
| 217 |
|
New S/R to copy tracer value from the OB location to narrow band region |
| 218 |
|
(width = same as overlap) beyond the OB: preserve OB value (needed for |
| 219 |
|
high-order scheme) which were overwritten by EXCH call if in overlap |
| 220 |
|
region. Applied to theta, salt, ptracer and advected seaice variables. |
| 221 |
|
With seaice: use maskInC in similar way as in gad_advection.F, |
| 222 |
|
in seaice_advection.F and advect.F. |
| 223 |
|
Update output of experiment seaice_obcs and change exp4 to test OBCS with |
| 224 |
|
4th order Adv.Scheme. for pasive salt and ptracer. |
| 225 |
|
o tools/genmake2: |
| 226 |
|
- when looking for an optfile (not specified), only search for Fortran compiler |
| 227 |
|
(remove sym-link test + setting of CPP + makedepend check + C-compiler check) |
| 228 |
|
- in get_fortran_c_namemangling: use "$CC $CFLAG" to compile C test programm |
| 229 |
|
(instead of "$MAKE genmake_test.o" since an existing Makefile can be wrong) |
| 230 |
|
- set default for CPP to "cpp -traditional -P" (was the case without specifying |
| 231 |
|
an optfile, but otherwise default was just "cpp". |
| 232 |
|
- comment out PLATFORM=$p_PLATFORM (no idea where p_PLATFORM comes from) |
| 233 |
|
- when "CC" is not set in the optfile, look for possible C compiler |
| 234 |
|
(instead of just setting CC=cc) |
| 235 |
|
- put back version v.169 modif (from 2008/02/15), was dropped in v.173 changes: |
| 236 |
|
=> add a ".F" to ".o" rule for source files with no "#include" (such as |
| 237 |
|
port_rand.F); otherwise, default ignores FOPTIM and just uses FFLAGS. |
| 238 |
|
- move gfortran before g77 in list of possible Fortran compiler (consistent |
| 239 |
|
with reference compiler switch from g77 to gfortran) |
| 240 |
|
o pkg/seaice: |
| 241 |
|
o remove SEAICE_OLD_AND_BAD_DISCRETIZATION-code, add a test to |
| 242 |
|
seaice_check that stops the model, when the CPP flag is used. |
| 243 |
|
o seaice_growth.F: make capping of ice volume conserving by returning the |
| 244 |
|
capped ice as freshwater to EmPmR. This does not change the verification |
| 245 |
|
experiments but will change runs where the capping is acutally active |
| 246 |
|
o change reference platform+compiler from faulks+g77 to baudelaire+gfortran |
| 247 |
|
(except the 2 fizhi test exp.); update output of exp.: bottom_ctrl_5x5, |
| 248 |
|
global_ocean.cs32x15 (icedyn, thsice, AD thsice), global_ocean_ebm (AD), |
| 249 |
|
lab_sea (fd, hb87, salt_plume), MLAdjust (3/5) and vermix (dd) |
| 250 |
|
o tools/genmake2: |
| 251 |
|
- merge genmake_warnings & genmake_errors into 1 log file: "genmake.log") |
| 252 |
|
- change DUMPSTATE default to false (can use "-ds" option to switch it to T) |
| 253 |
|
- clean previous genmake log files before anything else. |
| 254 |
|
- change Makefile to remove testreport log files (*.tr_log) and genmake log |
| 255 |
|
files only with "CLEAN" target (instead of "Clean"); |
| 256 |
|
o pkg/generic_advdiff: |
| 257 |
|
Change OBCS in multiDim advection (gad_advection.F): |
| 258 |
|
Use maskInC to prevent intermediate update of tracer-field in OB regions, |
| 259 |
|
thus keeping current-time-step OB values; and remove the call to OBCS_APPLY |
| 260 |
|
which were applying future time-step OB values. |
| 261 |
|
This way, all explicit calculations use current-time-step OB values |
| 262 |
|
(simpler and more consistent). |
| 263 |
|
Affects results when set-up uses time-varying OBC and multi-dim advection; |
| 264 |
|
update output of experiment seaice_obcs. |
| 265 |
|
o pkg/seaice: |
| 266 |
|
move OPTIONS (formerly locally defined) to SEAICE_OPTIONS.h: |
| 267 |
|
this includes SEAICE_GROWTH_LEGACY (+ SEAICE_DO_OPEN_WATER_GROWTH & |
| 268 |
|
SEAICE_OCN_MELT_ACT_ON_AREA ) and SEAICE_SOLVE4TEMP_LEGACY. |
| 269 |
|
Default (pkg/seaice/SEAICE_OPTIONS.h) is to define SEAICE_SOLVE4TEMP_LEGACY |
| 270 |
|
but leave SEAICE_GROWTH_LEGACY undef ; For now, keep verification exp |
| 271 |
|
unchanged (SEAICE_GROWTH_LEGACY defined in */code*/SEAICE_OPTIONS.h) |
| 272 |
|
o pkg aim (2010/10/26): |
| 273 |
|
- Introduce ALLOW_INSOLATION, moving toward a more general incoming solar radiation |
| 274 |
|
formulation with obliquity, eccentricity, ... as run-time parameters, |
| 275 |
|
- Introduce diagnostics for top-of-the-atmaophere incoming solar radiation and |
| 276 |
|
upward shortwave radition at the surface (all-sky and clear-sky). |
| 277 |
|
o pkg/obcs: |
| 278 |
|
- writing sections with globalFiles=T does not work: switch to tiled-files |
| 279 |
|
when writing sections for OBCS (change pkg/rw/write_rec.F accordingly). |
| 280 |
|
- fix index bug (OB_Ie -> OB_Iw_) in obcs_apply_uvice.F |
| 281 |
|
- apply OBCS_SEAICE_COMPUTE_UVICE to both uice and vice at all obcs |
| 282 |
|
|
| 283 |
|
checkpoint62m (2010/10/21) |
| 284 |
|
o pkg/seaice: vectorize seaice_solve4temp.F |
| 285 |
|
- move iterations loops outside of i/j-loops |
| 286 |
|
- requires 2D arrays for TB and effConduct |
| 287 |
|
- add 2D iceOrNot to replace HICE_ACTUAL.GT.0, as HICE_ACTUAL is |
| 288 |
|
modified in this routine |
| 289 |
|
- a little bit of clean up, adjusting indentation, etc. |
| 290 |
|
- replaced a few TEN**a by exp(a*log(10.)) in new code for better |
| 291 |
|
performance and '**TWO' by '**2' (because TWO is a float), |
| 292 |
|
replaced ONE, ZERO |
| 293 |
|
o eesupp/src: |
| 294 |
|
fix flow dir for exch1_z_RX_cube, exch1_uv_RX_cube & exch1_bg_RX_cube |
| 295 |
|
o pkg/seaice/seaice_growth.F |
| 296 |
|
Revision 1.89 of seaice_growth.F completes the overall october 2010 revision and |
| 297 |
|
merging process. The merged routine has two branches, and one activates the 'evolution' |
| 298 |
|
branch by setting SEAICE_GROWTH_LEGACY to undef (at the top of the routine for now). |
| 299 |
|
In summary, here is what the '10/10 revision' did consist of |
| 300 |
|
part 1: seaice_growth.F revisions 1.71 to 1.75 |
| 301 |
|
- Preliminary seaice_growth.F clean-up, keeping test results strictly unchanged. |
| 302 |
|
The emphasis here was on improving readability (e.g. by trying to use meaningful |
| 303 |
|
variable names) and modularity (e.g. by adding arrays to store heff etc. updates). |
| 304 |
|
part 1.5: |
| 305 |
|
- revising the lab_sea experiment to make it a more useful (i.e. stable |
| 306 |
|
and representative of an actual model run). This was done by generating a |
| 307 |
|
(winter) pickup to start from, switching back the main experiment dynamic |
| 308 |
|
solver from evp to lsr, having one of the experiments use the free drift solution. |
| 309 |
|
part 2: seaice_growth.F revisions 1.76 to 1.85 |
| 310 |
|
- Sorting out hard-coded constants that prevented heat and fresh water conservation. |
| 311 |
|
- Sorting out the various fields units and sign (set early on and reset at the end) |
| 312 |
|
to avoid numerous scalings/re-scalings. By v1.85 the basic unit for ice, heat and |
| 313 |
|
fresh water stock is 'effective ice meters', and >0 implies more ice/snow. |
| 314 |
|
- Using tendency terms arrays (d_HEFF* etc.) to alleviate remaining code locks, |
| 315 |
|
and simplifying logical expressions to further improve readibility. |
| 316 |
|
part 3: seaice_growth.F revisions 1.86 to 1.89 |
| 317 |
|
- Re-organize the blocks of code more logically. |
| 318 |
|
- Merging the '_if.F' pieces that add something new. |
| 319 |
|
o pkg/thsice: |
| 320 |
|
- vectorize thsice_solve4temp and thsice_get_exf. |
| 321 |
|
Problem: thsice_solve4temp has an unvectorizable statement that |
| 322 |
|
might also be problematic for TAF |
| 323 |
|
- thsice_get_exf: add the ALLOW_BULK_LARGEYEAGER04 CPP-flag. This |
| 324 |
|
does not change the verifcation experiment, but will change all |
| 325 |
|
experiments that have this flag undefined (not tested in verification); |
| 326 |
|
a little reorganization to save a few 2D fields and recomputations |
| 327 |
|
- 1st step of revisiting thsice adjoint (major recomp. sorted out) |
| 328 |
|
- 2nd step: fix some store dirs |
| 329 |
|
o seems to work somewhat, i.e. using ifort but not g77 |
| 330 |
|
o likely cause are incomplete init. or recomputation |
| 331 |
|
o start verification/global_ocean.cs32x15/input_ad.thsice |
| 332 |
|
o pkg/obcs: add support for exch2 |
| 333 |
|
- the position of the boundary and prescribed values are specified using |
| 334 |
|
a global domain with exch2 facets stacked |
| 335 |
|
- in x for N,S boundaries (like W2_mapIO=-1) |
| 336 |
|
- in y for E,W boundaries (so E,W boundaries do not overlap) |
| 337 |
|
o pkg/mdsio: add support for writing global sections with exch2 |
| 338 |
|
file layout is as for obcs described above: |
| 339 |
|
- for x-z sections, facets are stacked in x (like W2_mapIO=-1) |
| 340 |
|
- for y-z sections, facets are stacked in y |
| 341 |
|
o pkg/thsice: |
| 342 |
|
- change a few routines for better vectorization: thsice_albedo.F, |
| 343 |
|
thsice_step_temp.F, and in thsice_calc_thickn.F. Results are |
| 344 |
|
not affected, but in particular the latter routine has changed |
| 345 |
|
dramatically (not necessarily easier to read now). |
| 346 |
|
- re-arrange calculation in THSICE_SOLVE4TEMP: move outside of i,j loops |
| 347 |
|
iteration loop and call to S/R THSICE_GET_BULKF & THSICE_GET_EXF. |
| 348 |
|
o verification/lab_sea: |
| 349 |
|
- overhaul of the forward model experiments. The main motivation |
| 350 |
|
for this revision is that lab_sea has proven quite fragile (when |
| 351 |
|
changing the sea-ice code, and switching platforms) to a point |
| 352 |
|
when failures were very hard to attribute to specific code problems. |
| 353 |
|
- the suspected reason for this fragility is noise amplification by evp so, |
| 354 |
|
to start with, we decided to replace evp with lsr in the main experiment. |
| 355 |
|
- the input.lsr experiment thus became obsolete and was removed. |
| 356 |
|
- it was replaced with the input.fd experiment that uses an |
| 357 |
|
analytical "free drift" solution as the ice velocity field. |
| 358 |
|
- Taken together, input (lsr), input.fd (free drift) and input.hb87 (evp) |
| 359 |
|
will hopefully allow us to discriminate problems due to dynamic solvers |
| 360 |
|
from problems due to the thermodynamic code. input.salt_plume (lsr) allows |
| 361 |
|
us to test additional options, along with the salt plume package. |
| 362 |
|
- Finally a pickup was generated (at the end of a one year run) to |
| 363 |
|
make the tests more representative of an actual model run. Previously |
| 364 |
|
the tests starting point was a 1m thickness of ice everywhere. |
| 365 |
|
o pkg/seaice: |
| 366 |
|
- introducing S\R seaice_freedrift.F that compute the |
| 367 |
|
analytical solution to an approximate ice momentum |
| 368 |
|
equation (i.e. the simple stress/coriolis balance). |
| 369 |
|
- define SEAICE_ALLOW_FREEDRIFT to trigger the computation |
| 370 |
|
of the free drift velocity field, and enable its use. |
| 371 |
|
To actually use the free drift solution as the model ice |
| 372 |
|
velocity field set SEAICEuseFREEDRIFT to .TRUE. in data.seaice. |
| 373 |
|
o move "useOASIS" from PARAMS.h to EEPARAMS.h and read it from "eedata" |
| 374 |
|
|
| 375 |
|
checkpoint62l (2010/10/04) |
| 376 |
|
o pkg/seaice: |
| 377 |
|
-fix the USE_ORIGINAL_SBI version (for variable freezing point); |
| 378 |
|
-use "celsius2K" in SEAICE_VARIABLE_FREEZING_POINT and in new |
| 379 |
|
version (with USE_ORIGINAL_SBI undef); |
| 380 |
|
o add hooks for new packages OASIS, the package will follow |
| 381 |
|
(hooks exclude the seaice pkg for now) |
| 382 |
|
o pkg/autodiff: ALLOW_AUTODIFF_WHTAPEIO |
| 383 |
|
Option to use mdsio_read_whalos.F/mdsio_write_whalos.F, |
| 384 |
|
rather than mdsio_readvector.F/mdsio_writevector.F, within |
| 385 |
|
adread_adwrite.F to read/write adjoint-related tapes to disk. |
| 386 |
|
o eesupp and pkg/mdsio: ALLOW_WHIO |
| 387 |
|
Option to write/read 2D fields to files including tiles halos. |
| 388 |
|
The main purpose is for adjoint related "tape I/O". |
| 389 |
|
The secondary purpose is for debugging phases. |
| 390 |
A word on those rather specifically intended purposes. |
A word on those rather specifically intended purposes. |
| 391 |
The code was meant to be stripped down to the bare minimum. |
The code was meant to be stripped down to the bare minimum. |
| 392 |
Typically, there is a unique and basic mapping to the global |
Typically, there is a unique and basic mapping to the global |
| 393 |
buffer (tile by tile, then proc by proc, then level by level), and |
buffer (tile by tile, then proc by proc, then level by level), and |
| 394 |
byteswaps and meta-files are omitted. Indeed the (*_WH*) code is not |
byteswaps and meta-files are omitted. Indeed the (*_WH*) code is not |
| 395 |
meant to generate user friendly outputs of various shapes and forms. |
meant to generate user friendly outputs of various shapes and forms. |
| 396 |
Rather it is intended, for the knowledgeabale developers, |
Rather it is intended, for the knowledgeabale developers, |
| 397 |
to be largely independent of the many mdsio options, to allow |
to be largely independent of the many mdsio options, to allow |
| 398 |
outputs of tiles+halos in debugging phases and in adjoint runs, |
outputs of tiles+halos in debugging phases and in adjoint runs, |
| 399 |
and to allow some flexibility in tuning (adoint) I/O depending on each |
and to allow some flexibility in tuning (adoint) I/O depending on each |
| 400 |
file system behavior. With regard to tuning, most of the potential knobs |
file system behavior. With regard to tuning, most of the potential knobs |
| 401 |
(to leave files open, to allocate the I/O indep. of usesinglecpuio, |
(to leave files open, to allocate the I/O indep. of usesinglecpuio, |
| 402 |
to specify directories, to specify the typical size of I/O calls, etc.) |
to specify directories, to specify the typical size of I/O calls, etc.) |
| 403 |
are not yet included, and will be added if they prove useful. |
are not yet included, and will be added if they prove useful. |
| 690 |
but keep the former optfile for few weeks for smooth transition |
but keep the former optfile for few weeks for smooth transition |
| 691 |
o pkg/ecco |
o pkg/ecco |
| 692 |
- allow user defined cost function contributions |
- allow user defined cost function contributions |
| 693 |
CPP: ALLOW_USERCOST_CONTRIBUTION |
CPP: ALLOW_USERCOST_CONTRIBUTION |
| 694 |
USAGE: compute contributions in cost_usercost_all.F |
USAGE: compute contributions in cost_usercost_all.F |
| 695 |
- note: this is identical to what ALLOW_GENCOST_CONTRIBUTION once was. |
- note: this is identical to what ALLOW_GENCOST_CONTRIBUTION once was. |
| 696 |
the two options are complementary. |
the two options are complementary. |
| 697 |
o pkg/smooth |
o pkg/smooth |
Parent Directory
| 
Revision Log
| 
Revision Graph
| 
Patch