/[MITgcm]/MITgcm/doc/tag-index
ViewVC logotype

Diff of /MITgcm/doc/tag-index

Parent Directory Parent Directory | Revision Log Revision Log | View Revision Graph Revision Graph | View Patch Patch

revision 1.1367 by mlosch, Wed Jan 19 13:52:14 2011 UTC revision 1.1508 by jmc, Thu Sep 29 17:31:02 2011 UTC
# Line 4  $Name$ Line 4  $Name$
4      Notes on tags used in MITgcmUV      Notes on tags used in MITgcmUV
5      ==============================      ==============================
6    
7    o tools/build_options:
8      - after merging some MPI and/or multi-threads specific optfiles so that
9        the modified default optfile support both built (using genmake2 -mpi
10        for the MPI built, using genmake2 -omp for OpenMP built),
11        move now obsolete optfiles to tools/build_options/unsupported/
12       (add copy of email sent to support list in doc/optfiles_changes.txt)
13    
14    checkpoint63c (2011/09/28)
15    o pkg/ebm:
16      - fix ebm_zonalmean.F for several procs in Ydir (nPy > 1); update output of
17        test exp. global_ocean_ebm (affects machine truncation, 12 digits for cg2d)
18    o pkg/seaice (SItracer pickup)
19    * Simple code for pickup of SItracer fields (#define ALLOW_SITRACER)
20      To be done:
21      - thorough testing of missing fields in S/R seaice_read_pickup
22      - thorough matching between generic tracers and specific fields used
23        (e.g. volume-weighted age, area-weighted age, salinity, ...)
24    o pkg/grdchk:
25      - all procs calculate and print finite-difference gradient (for testreport)
26      - change description output used by testreport;
27      - AD testing: also compare and report Finite-difference gradient;
28        => update all output_adm*.txt
29    o test exp. global_ocean_ebm:
30      - switch back useRealFreshWaterFlux (no problem anymore with run-off);
31        and turn on exactConserve; update (fwd) output.
32    o pkg/seaice (LSR):
33      - print to STDOUT number of solver iterations at SEAICE_monFreq interval.
34      - modify solver to deal cleanly with u/v switch at face edges (CS-grid):
35        * identical results with non-CS topology (but save up to half of EXCH calls)
36        * no much changes in cs-32 verification tests since no ice at face edges
37         (differences due to convergence criteria, since nb of iterations changed)
38        * expect some differences on lat-lon-cap grid.
39        update FWD results of global_ocean.cs32x15 test exp. (icedyn & seaice)
40    o pkg/exch2:
41      - add new CPP option to disable Cumul-Sum matrix (to save storage when using
42        many tiles); until alternative coding is added, just add a stop.
43    o pkg/kpp:
44      - move ptracer/salt correction to KPP diffusivity from calc_3d_diffusivity.F
45        to S/R KPP_CALC_DIFF_PTR (+ add tracer-index to argument list).
46    o optfiles:
47      - merged few MPI specific (*+mpi*) and multi-threads specific (*+mth) optfiles
48        into corresponding standard one, which now support different build:
49        (when using "genmake2 -mpi" for MPI built, "genmake2 -omp" for OpenMP built)
50      - add "darwin_amd64_gfortran" optfile (initially from Constantinos, tested by Gael)
51    o pkg/obcs (seaice):
52      - add missing EXCH for uIce,vIce after applying OBCS;
53      - remove S/R EXF_SWAPFFIELDS_X/Y , disable S/R EXF_SET_OBCS_X/Y by adding
54        argument = number of levels to S/R EXF_SWAPFFIELDS_XZ/YZ and S/R
55        EXF_SET_OBCS_XZ/YZ ; update obcs_exf_load.F;
56      - fix EXF_SET_OBCS_XZ/YZ for multi-threaded (assuming loaded arrays are shared)
57      - verification exp. seaice_obcs: split domain in 2 tiles ; add SIZE.h_mpi
58        and eedata.mth for MPI and mutlti-threaded testing.
59    o pkg/flt:
60      - make FLT EXCH working with multi-threads (and MPI + multi-threads);
61      - use buffer for floats IO (trajectories & profiles); size set in FLT_SIZE.h;
62      - fix pickup and trajectories + profiles output with multi-threads;
63      - add parameter to select group of fields to write along trajectory and
64        profile (default unchanged);
65      - only call PORT_RAND* function when flt_noise <> 0;
66      - add multi-threaded test for verification exp. flt_example.
67    o eesupp/src:
68      - fix S/R ALL_PROC_DIE for MPI+multi-threads.
69      - USE_OMP_THREADING: check that number of threads (from eedata) matches
70        actual number (from env.var. setting).
71    o pkg/ebm:
72      - fix multi-threading in ebm_area_t.F ; clean-up EBM.h header file.
73      - add multi-threaded test for verification exp. global_ocean_ebm
74    
75    checkpoint63b (2011/08/29)
76    o pkg/ebm:
77      - fix sign of runoff;
78      - update directly forcing fields (Qnet,EmPmR) to use standard
79        external_forcing_surf.F (remove ebm_forcing_surf.F);
80      - avoid loading forcing fields a 2nd time (1rst time from ini_forcing.F);
81      - set either tauThetaClimRelax (in data) or tauThetaZonRelax (in data.ebm)
82        and apply relaxation towards corresponding SST field (2-D map or ZonAver).
83      - update global_ocean_ebm parameters file data and update output (fwd+ad).
84    o verification/matrix_example:
85      - split domain in 8 tiles (2x4); add SIZE.h_mpi and eedata.mth for
86        MPI and mutlti-threaded testing.
87    o pkg/matrix:
88      - change name of S/R MATRIX_INIT to S/R MATRIX_INIT_VARIA ;
89      - fix for multi-tiles (add bi,bj indices to exp/impMatrixCounter);
90      - fix for multi-threaded (remove DATA statement);
91      - use standard DIFFERENT_MULTIPLE function;
92      - add MATRIX_OPTIONS.h
93    o verification/matrix_example:
94       To allow to split the domain in several tiles, change the domain size
95        from 31x31 to 32x32 by adding 1 row (on Southern side) + 1 column (on
96        Western side) of land points.
97    o (*_)write_pickup:
98      - fix few bugs in mypackage_read/write_pickup.F
99      - in MDS_WR_METAFILES calls, replace "myTime" by a local copy into
100        array of length 1 (was caught by compiler syntax-check).
101    o pkg/diagnostics (velocity potential solver):
102      - fix initialisation of main diagonal matrix (for overlap values)
103    o tools/build_options:
104      - towards single optfile for MPI and non-MPI built: merge few +mpi optfile
105        into previously non MPI one (linux_am64 pgf77,open64,pathf90,ifort11)
106      - linux_amd64_gfortran: merge +mpi_generic version into standard optfile:
107        * move IEEE flags to DEVEL section; take IEEE flags from +mpi_generic ;
108        * FAST case: for now, keep -O3 with ini_masks_etc.F in NOOPTFILES list;
109        and use "-devel" testreport option for standard testing on baudelaire.
110    o pkg/grdchk:
111      - Assume that all gradient-check position are within the same tile
112        corresponding to global indices (iGloTile,jGloTile) and that
113        positions (i,j indices or nbeg) are local to this tile;
114      - set tile local-indices and proc. number according to tile global-indices;
115      - remove grdchkwhichproc from namelist (retired param);
116      - send adjoint gradient to all procs so that all procs print it to STDOUT;
117      - switch on MPI test for global_ocean.90x40x15 AD test experiment.
118    o adding package bbl (Bottom Boundary Layer)
119      description in MITgcm/pkg/bbl/bbl_description.tex
120      example/test experiment in MITgcm_contrib/bbl
121    o adjoint verifications
122      - changed 5 experiments to multi-tiles
123        * global_ocean.90x40x15
124        * obcs_ctrl
125        * offline_exf_seaice
126        * tutorial_dic_adjoffline
127        * tutorial_tracer_adjsens
128        changed reference output for those that crossed tile boundaries
129        (changed choices of gradient check positions)
130      - with same tiling, switch on MPI test for the last 4 (add SIZE.h_mpi)
131      - temporary bug fix in grdchk_loc.F for obcs
132    o genmake2 & testreport
133      - add target "obj" (just for OBJFILES files) to Makefile;
134      - add option to specify FEXTRAFLAGS on command line.
135    o pkg/mypackage:
136      - remove bi,bj arguments from argument list of S/R MYPACKAGE_CALC_RHS
137    
138    checkpoint63a (2011/08/02)
139    o pkg/diagnostics (velocity potential):
140      - fix velocity-potential calculation when using OBCS: balance net inflow
141        at each level (new S/R OBCS_DIAG_BALANCE).
142      - fix some of the problems with OBCS in stream-function calculation.
143    o pkg/exf/exf_readparms.F
144      - make treatment of optional namelists (EXF_NML_04, EXF_NML_SGRUNOFF,
145        EXF_NML_OBCS) more consistent, so that you need not have these lists
146        in data.exf when they are not used.
147    o optim/optim_readdata.F:
148      - add some output about obcs variables that might help debugging
149      - add a check about nvarlength
150    o pkg/ctrl: once again, fix computation of endrec in s/r ctrl_init_rec
151    o pkg/diagnostics (velocity potential):
152      - fix mask for OBCS (still problems in stream-function with OBCS);
153      - add specific parameter (default = main code CG2D params) for solver;
154      - in case of poor convergence, use solution corresponding to lowest residual.
155    o pkg/seaice/seaice_growth.F: small changes that permit vectorizable
156      adjoint code. In general statements like tmpscal=fld(i,j) are problematic
157      and should be avoided where possible.
158    o model/src (rStar):
159      - include r* effect on reference-state (PhiRef) in totPhiHyd
160        (so that it's really the potential anomaly at the cell center);
161        affect solution only if z* with full pressure in EOS.
162      - test-exp global_ocean.90x40x15: reduce viscA4 (was not stable) and
163        update output (because of z* in totPhiHyd and viscA4 changes);
164      - keep the previous diagnostics as "PHIHYDcR" (closer to potential anomaly
165        @ constant r) until interpolation at constant r is implemented.
166    o pkg/diagnostics
167      - fix RCENTER diagnostics in case cell center is not at interface middle
168    o add code for TEOS-10 (www.teos-10.org, McDougall et al. 2011). Use
169      this eos with eosType = 'TEOS10', in data (PARM01). This eos implies
170      that THETA and SALT are "conservative temperature" and "absolute
171      salinity"
172    o pkg/ctrl
173      - fix computation of endrec in s/r ctrl_init_rec
174      - ctrl_pack/unpack_*: remove unused variables and some obsolete code
175    o pkg/seaice/seaice_growth.F
176      - fix recomputation with FENTY_AREA_EXPANSION_CONTRACTION code,
177        remove then obsolete STORE directives
178      - fix STORE directives for tice and tices, so that only the
179        appropriate part of the fields are stored (and not the entire
180        field); update global_ocean.cs32x15.seaice AD output.
181      - cosmetic changes: adjust CPP flags and indentation for better
182        legibilty
183    o pkg/autodiff: add mnc-output of adBottomDragFld (code contributed by
184      Chris Wilson)
185    o pkg/gmredi:
186       Add Sub-Meso Eddies parameterisation from Baylor (for now, disabled with TAF),
187       and test it in verification exp. "front_relax.bvp" (update output.bvp.txt)
188    
189    checkpoint63  (2011/07/11)
190    o pkg/exch2:
191     - rename+move: nTiles in W2_EXCH2_PARAMS.h --> exch2_nTiles in W2_EXCH2_TOPOLOGY.h
192     - exch2 version of Cumul-Sum routine: S/R W2_SET_MAP_CUMSUM set up the
193       (2*nTiles)x(nTiles) Cumul-Sum matrix and Master Proc collects all tiles X,Y
194       increments, calculates Cumul-Sum @ tile origin and sends results to each tile.
195    o pkg/diagnostics:
196     - deal with pairs of post-processed diags which are computed together; in
197       particular, re-order list of diags within 1 file to save 2nd computation.
198     - add S/R to compute velocity stream-function (called after velocity-potential
199       calculation) from divergence free transport -> new diagnostics "PsiVEL".
200     - add parameter to select grid-point location where PsiVEL == 0.
201    o eesupp/src:
202       new S/R to calculate cumulated sum on tiled array, corner grid-cell location
203       (specific S/R needed with exch2 topology - for now, just print an error msg)
204    o pkg/shelfice
205     - Implement friction velocity-dependent transfer coefficients following
206       Holland and Jenkins, JPO, 1999
207       Original code by M. Losch with small modifs.
208       Enable with CPP option SHI_ALLOW_GAMMAFRICT
209       and runtime flag SHELFICEuseGammaFrict
210    o verification lab_sea & global_ocean.cs32x15:
211     - use tools/xmakedepend (instead of cyrus_makedepend, in genmake_local):
212       does not miss some dependencies (but is slower) and also works with many files
213       (default makedepend limitation for AD built).
214     - use cyrus_makedepend in adjustment.128x64x1 (so that it still tested).
215    o pkg/diagnostics:
216     - add run-time parameter "useMissingValue" (def=False) to fill land-point
217        (i.e., where mask=0) with MissingValue ; used only in MNC output file.
218       This replaces CPP-option DIAGNOSTICS_MISSING_VALUE.
219     - Extend the use of "missing_value" setting from data.diagnostics for the case
220       of counter-diagnostics (where fraction-weight is zero). This applies both to
221       MNC and MDS output files.
222    o pkg/seaice/seaice_growth.F:
223      - add 'a priori' cap of latent heat flux based on actual mass considerations
224        when SEAICE_ADD_SUBLIMATION_TO_FWBUDGET is used in conjunction with the
225        non-legacy seaice_solve4temp.
226      - add diagnostic of the residual sublimation freshwater flux (the freshwater
227        flux taken from the ocean if the total actual latent heat fluxes exceed
228        the maximum)
229      - remove the CPP switch around the *_FWbySublim unit change and move diagnostic
230       fill for SIatmQnt, SIfwSubl, SIatmFW to the end of seaice_growth.F.
231      - add diagnostic of the actual sublimation freshwater flux (that is 0. ifndef
232       SEAICE_ADD_SUBLIMATION_TO_FWBUDGET) and of the latent heat flux (evap+sublim).
233      - bug fix : frWtrAtm needs a_FWbySublim ifdef SEAICE_ADD_SUBLIMATION_TO_FWBUDGET.
234      - bug fix : missing part to Martin's r_FWbySublim pathological case fix.
235    o verification 1D_ocean_ice_column:
236       made model domain truly 1x1x23 instead of 5x5x23 with only center cell (3,3,:) wet
237       update results
238    o pkg/diagnostics:
239     - Implement setting of "Post-Processed" diagnostics (corresponding to gdiag(5)='P')
240       which are not filled-up but computed from other diags ; In this case, the mate
241       diag points to the primary (filled-up) diag to used for post processing.
242     - Update Velocity-potential computation using post-processed diag framework.
243     ==> Change the ordering of fields and time-periods in MDS output file
244       when using periodic averaging: now writes one time record with the full set
245       of fields for this time period, then the next time reccord (it used to be
246       one field with the full set of time records then the next field).
247         Note: this is now similar to NetCDF (MNC) output file. <==
248    
249    checkpoint62z (2011/06/21)
250    o adjoint monitor:
251     - Add adjoint monitor output for packages exf, seaice
252       and update relevant testreport output
253    o pkg/seaice:
254     - re-order diags inits so that they appear in order in available_diagnostics.log
255     - added diags: hsnow thermo. rates of change, area preceeding ridiging process,
256       area/heff/hsnow preceeding thermodynamic growth/melt.
257     - modified diag: SIdA; output actual effect, as the description states.
258     - modified description: SIqneti, SIqneto, SIuheff, SIvheff, SIdAb*, HSNOW.
259     - retired redundant diags: SIyneg, SIfice, SIsnwice, SIthdgrh.
260       (SIyneg->SIdHbOCN; SIfice->SIdHbATC; SIsnwice->SIdHbFLO;
261        and SIthdgrh->SIdHbOCN+SIdHbATC+SIdHbATO)
262     - modified diags (2) : switch all SIdA*, SIdH*, SIdS* from delta to rates.
263     - added diagnostics for seaice generic tracers (SItr*).
264     - added SItrUnit and SItrNameLong run time param (for SItr* diags).
265     - in diag names, replaced 'PrTh' abbrev. of 'preceeding thermo' with 'PT'.
266     I. Fenty: 18-Jun-2011:
267     Minor changes to seaice package:
268     1) Retired old variables (A22, SEAICE_lhsublim, areaMax, areaMin, hiceMin) and
269        added some new ones (SEAICE_area_reg, SEAICE_hice_reg, SEAICE_area_floor)
270       - Differentiated "regularization variables" from "floor variables"
271        * areaMin became SEAICE_area_reg (old A22) and SEAICE_area_floor
272        * hiceMin became SEAICE_hice_reg (old hiceMin)
273          (with _reg meaning regularization variable)
274       - SEAICE_lhSublim becomes lhSublim, the sum of SEAICE_lhEvap and SEAICE_lhFusion
275        so as to ensure energy conservation when going between phases
276      - A22 was not used anywhere
277     2) Changed regularization procedure for heffActual and hsnowActual to ensure
278        well-boundedness and smooth adjoint in seaice_growth.F
279     3) Fixed a bug where seaice_solve4temp would not recognize ice-free grid cells
280        because the old regularization always set heffActual >= 0.05 cm
281     4) Changed the model so that the default behavior is to put a small (10^-5) "floor"
282        on AREA when HEFF > 0.
283      - went from requiring ALLOW_PRECLUDE_INFINITESIMAL_AREA to be defined to
284        requiring that DISABLE_AREA_FLOOR *not* be defined
285     Affected verifications:
286      - 1D_ocean_ice_column (output.txt output_adm.txt)
287      - global_ocean.cs32x15 (output.seaice.txt output_adm.seaice.txt)
288    o pkg/diagnostics:
289      - move MNC code from diagnostics_out.F into new diagnostics_mnc_out.F (2 S/R).
290      - rename S/R GETDIAG to DIAGNOSTICS_GET_DIAG and change type of 1rst argument
291        (was _RL, now integer) with option = 0 to retrieve all levels.
292      - add pieces of code to compute velocity potential at each level (new diagnostics
293        PhiVEL, works only when UVELMASS & VVELMASS are also turned on);
294        for now, using same params as CG2D (target-resisual, max Nb of iter ...).
295      - fix few issues with counter diagnostics in diagnostics_setdiag.F &
296        diagstats_setdiag.F (if diagnostic counter is itself a counter diag)
297        and in diagnostics_fract_fill.F (do not need a valid pointer for the
298        diagnostics counter-mate); check that definition of counter-diagnostic
299        includes a valid counter-mate (in diagnostics_set_levels.F).
300    o model/src (rStar):
301      - add rStar scaling factor in diags_phi_hyd.F (similar to calc_grad_phi_hyd.F);
302        affect solution only if z* with full pressure in EOS;
303        update output of test-exp global_ocean.90x40x15.
304    o pkg/seaice:
305      - introduce siEps=1e-5 parameter
306      - use areaMax in AREA update (part 4), consistent with ridging step (part 2.5).
307      - re-organize SEAICE_ADD_SUBLIMATION_TO_FWBUDGET code.
308      - introduce ALLOW_SITRACER and seaice_tracer_phys.F to handle generic seaice tracer.
309        For now it covers, and was tested for, ice salinity and age (attached to HEFF),
310        as well as ice cover age and ridging index (attached to AREA).
311        For now the run time parameters (in data.seaice/SEAICE_PARM03) are
312        SItrName (e.g. 'age') and SItrMate ('HEFF', which is the default, or 'AREA').
313        Up to SItrMaxNum=3 tracers can be used simultaneously.
314    o pkg/salt_plume:
315      -introduce SPsalFRAC run time parameter that sets the fraction of the
316       salt by-product of ice growth (SPsalFRAC*(1-SIsalFRAC)*salt typically)
317       that gets vertically redistributed (fluxed down) by pkg/salt_plume.
318    o model debug options:
319      - Switch reading of debugMode from S/R INI_PARMS (parameter file "data", stored
320        in PARAMS.h) to S/R EESET_PARMS (parameter file "eedata", stored in EEPARAMS.h).
321      - Refine debug-msg printing selection (3 new levels):
322        debugMode  : controls printing of debug messages (sequence of S/R calls).
323        debugLevel : controls printing of algorithm-relevant auxiliary/debug messages
324         debLevZero =  0 ; /* level of disabled aux. msg printing */
325         debLevA =  1 ; /* level of minimum  aux. msg printing */
326         debLevB =  2 ; /* level of low aux. print (report read-file opening)*/
327         debLevC =  3 ; /* level of moderate debug prt (most pkgs debug msg) */
328         debLevD =  4 ; /* level of enhanced debug prt (add DEBUG_STATS prt) */
329         debLevE =  5 ; /* level of extensive debug printing */
330        New debugLevel default: with default debugMode value (=F),
331          = 2 (debLevB) or = 1 (debLevA) in case ALLOW_AUTODIFF is defined.;
332          and if debugMode=T, default debugLevel = 4 (debLevD).
333        Relation with previous settings:
334         debLevB (=2) corresponds to former debLevA (=1)
335         debLevD (=4) corresponds to former debLevB (=2)
336        Note:
337         a) if debugLevel=0, skip WRITE_GRID call when nIter0 > 0 (same as before);
338         b) default debugLevel=1 is convenient in AD runs (avoid many file opening msgs);
339            to recover previous default, set debugLevel=2 (e.g. AD-test hs94.1x64x5)
340      - Add new parameter (printResidualFreq, read from PARM02 namelist) to print
341        detailed CG solvers convergence; off (=0) by default, except if debugLevel=5.
342    o pkg/obcs:
343      - change OBCSfixTopo default from True to False (should not change the results);
344        keep OBCSfixTopo=T in verification-exp. seaice_obcs.
345    o pkg/thsice:
346      - fix bug in thsice_calc_thickn.F (was introduced in version 1.18) in
347        2nd layer seaice sublimation (unlikely to happen)
348      - use Interior mask for OBCS in thsice_advection.F (similar to what is
349        done in gad_advection & seaice_advection).
350    o adjoint:
351      bug fix spotted by Gael Forget:
352      seaice_advection and thsice_advection still had
353      legacy CPP options to avoid CS corner updates for adjoint
354    o pkg/seaice:
355      => Change some of the defaults (2011/05/29):
356         - change SEAICEuseFluxForm default (from False to True)
357         - change DIFF1 default (from 0.004 m/s to 0. m/s)
358         And print warnings if DIFF1 is not set in data.seaice and centered
359          avection scheme is used without any diffusion.
360      also:
361      - fix seaice_diffusion calls in default AdvScheme part of seaice_advdiff.F;
362      - set SEAICEdiffKh[] default in the same way as SEAICEadvSch[]
363      - stop if DIFF1<>0 with SEAICEadvScheme <> 2, if DIFF1<>0 with any
364        other non-zero SEAICEdiffKh[] diffusivity, if SEAICEadvSalt=T with
365        #undef SEAICE_VARIABLE_SALINITY, if SEAICEadvAge=T with #undef SEAICE_AGE.
366      update results of test exp.:
367         lab_sea: output.fd (switch to AdvScheme=33 with diffKh=20)
368             output.hb87, output_adm.evp (SEAICEuseFluxForm=T and diffKh=200)
369             output_adm.noseaicedyn (SEAICEuseFluxForm=T)
370         offline_exf_seaice: output.seaicetd, output_adm (SEAICEuseFluxForm=T)
371         seaice_obcs: output (SEAICEuseFluxForm=T)
372      and prescribe old default for SEAICEuseFluxForm (=F) and DIFF1 (=0.004)
373          for standard fwd & AD lab_sea tests (to keep the same output).
374    o pkg/seaice/seaice_growth.F:
375      - use Ian Fenty's capping formulas for actual ice
376      thickness thoughout EVOLUTION branch.
377      - put treatment pathological case #2) in CPP brackets
378      (ALLOW_AVOID_INFINITESIMAL_AREA, undef by default).
379      - update global_ocean.cs32x15 and 1D_ocean_ice_colum results accordingly.
380    o pkg/seaice:
381      - seaice_diffusion.F : remove the min(DX,DY) factor permanently;
382      pass the Laplacian diffusivity (in m^2/s) diffKh as a parameter;
383      also pass fac as a param, which can be 1. (to update the tendency
384      field) or the time step (to update the field itself).
385      - seaice_advdiff.F: for multidim, call seaice_diffusion with
386      e.g. SEAICEdiffKhHeff and with fac=1; for single-dim, call
387      seaice_diffusion with e.g. SEAICEdiffKhHeff and with fac=time step.
388      - SEAICE_PARAMS.h, seaice_readparms.F, seaice_check.F :
389      hooks to handle SEAICEdiffKhHeff, SEAICEdiffKhArea, etc.
390    o pkg/seaice:
391      diffusion with centered advection scheme (diffus.F & advect.F):
392      - simplify S/R DIFFUS (put deltaT & DIFF1 factor outside, move EXCH outside)
393        to just compute the laplacian of input-field.
394      - unroll k loop (harmonic,biharmonic diff) in advect.F to save 1 exch call.
395      - fix flux-form version in diffus.F to conserve tracer.
396      update output of lab_sea.fd & lab_sea.hb87 (1rst set of modif affects
397       machine truncation) and lab_sea.salt_plume (conservative flux-form modif)
398    
399    checkpoint62y (2011/05/26)
400    o pkg/seaice:
401     - seaice_growth.F : in EVOLUTION branch, always include
402       a_QSWbyATM_cover in qnet (as it should be).
403     - verification/global_ocean.cs32x15 : update results accordingly
404       (output.seaice.txt output_adm.seaice.txt).
405    o pkg/seaice + obcs:
406     - use maskInW & maskInS in multi-dim advection and biharmonic diffusion:
407       equivalent of assuming no tracer gradient outside the OB region.
408       update output of seaice_obcs (uses DIFF1 <>0 => biharmonic diffusion)
409     - remove calls to OBCS_COPY_TRACER after EXCH calls (no longer needed
410       after changes above);
411     - obcs+seaice cleaning: apply seaice-tracer OB values to tracer field
412       only at OB location (instead of over a band of OLx width).
413    o pkg/seaice:
414     - fix wrong indices in old (non Flux-Form) diffusion default S/R diffus.F
415       (dxG(I+1, ...) changed to dxG(I, ...), no impact on Lat-Lon grid)
416    o pkg/obcs:
417      - split header file "OBCS.h" into 4 separated files:
418        OBCS_PARAMS.h, OBCS_GRID.h, OBCS_FIELDS.h & OBCS_SEAICE.h
419       Note: OBCS_PARAMS.h & OBCS_GRID.h can be included without OBCS_OPTIONS.h
420         (contain no conditional statement which depends on OBCS options).
421      - move "useOBCSYearlyFields" from "data.obcs" (hold in OBCS_PARAMS.h)
422        to parameter file "data.exf", namelist EXF_NML_OBCS (in EXF_PARAM.h).
423    o ctrl, cost, grdchk & autodiff pkgs:
424      - make options file more standard (like other {PKG}_OPTIONS.h file but
425        with AD_CONFIG.h); clean-up some of the included header files.
426    o pkg/obcs:
427      - do not mask tracer when applying OB (tracer == 0 could generate problems)
428      - apply tracer OB values to tracer field only at OB location (instead of
429        over a band of OLx width)
430      - clean-up obcs_apply_w.F & obcs_apply_eta.F (since less calls to these S/R)
431      - remove OBCS_MULTIDIM_OLD_VERSION code in gad_advection.F and remove
432        obcs_apply_tloc.F & obcs_apply_sloc.F
433    o obcs in model/src:
434      - call OBCS_APPLY_W only if NonHydrostatic.
435      - call OBCS_APPLY_ETA only if NonLinFreeSurf.
436      - reset dEtaHdt and wVel to zero outside OB interior region ;
437      - leave tracer unchanged outside OB interior region: This has no effect
438        on the solution but just to prevent unrealistic tracer value outside OB.
439      - zero out velocity outside OB interior region: This has no effect on
440        the solution but just to prevent unrealistic velocity outside OB.
441    o obcs and non-lin free-surf:
442       move call to OBCS_APPLY_R_STAR from update_r_star.F to calc_r_star.F
443       and call to OBCS_APPLY_SURF_DR from update_surf_dr.F to calc_surf_dr.F
444       + add argument "etaFld" to OBCS_APPLY_R_STAR & OBCS_APPLY_SURF_DR:
445       this fixes missing EXCH + get consistent eta OB value (+ fix restart).
446       update results of test-exp. exp4(nlfs).
447    o exp4 (nlfs) test-exp:
448       use OBC for Eta at Western & Eastern OB (to test prescribing Eta
449       with NonLin-FreeSurf); update output.nlfs.txt
450    o obcs and 2d/3d solver norm:
451       use interior masks (maskInC) to cancel out cg2d & cg3d matrix coeff at OB
452       and outside OB interior region (will allow to place OB further inside the
453       domain). Depending on the domain, can affects cg2d-Norm (previous
454       cancellation outside OB was only partial), e.g. in exp4 & obcs_ctrl,
455       and cg3d-Norm (previous Norm was computed before canceling OB terms),
456       e.g. in tutorial_plume_on_slope.
457      - update output of exp.: exp4 (x2), obcs_ctrl and tutorial_plume_on_slope.
458    o pkg rbcs:
459      - split RBCS.h into 3 files: RBCS_SIZE.h, RBCS_PARAMS.h & RBCS_FIELDS.h
460      - add capability to apply relaxation to horizontal velocity uVel & vVel.
461    o pkg/grdchk: add gradient check for control variable xx_shifwflx
462    o pkg/shelfice/shelfice_init_fixed.F
463      fix CPP-flag combinations
464    o pkg/seaice: Merge iceage code of Pierre Rampal
465      * update/change/correct ice age calculations
466      * add age updates for pathological cases
467      Probably needs more testing...
468    
469    checkpoint62x (2011/05/13)
470    o pkg/mom_vecinv:
471       high-order vorticity scheme: use interior masks (maskInW & S)
472       to zero-out (outside OB) gradient of vorticity.
473    o verification exp.:
474       turn on "useEnergyConservingCoriolis" (better discretisation) in exp.
475       hs94.128x64x5 and seaice_obcs ; update results.
476    o pkg/obcs:
477      fix bug in OBeta for YZ sections in obcs_exf_load.F
478    o pkg/ctrl, pkg/shelfice, and optim:
479      - add new control variable xx_shifwflx (fresh water flux underneath ice
480        shelves). This is almost as tedious as obcs-ctrl, because the
481        variables needs its own mask.
482      - add a cost function term that penalizes the deviation of xx_shifwflx
483        from the first guess (zero) in analogy to other control variables
484    o pkg/shelfice:
485      - start basic adjoint
486    o pkg/ctrl:
487      - remove extra 2D field "for good measure" when initializing in
488        ctrl_set_globfld_[xy/xz/yz].F
489      - replace hard coded precision 32 (a bug) in ctrl_map_ini_gen by ctrlprec
490      - improve error messages in ctrl_set_unpack_*
491    o OBC implementation in momentum:
492       use interior masks (maskInC,W & S) in momentum S/R:
493       - multiply divergence, tension and some del^i.u or del^j.v by maskInC
494       - multiply del2u,del2v by maskInW/S (biharmonic viscosity)
495       - multiply horiz. grad of vorticity by maskInW/S (in Leith scheme)
496       - multiply gradient of wVel by maskInW/S in del2w calculation (NonHyd)
497       Motivation: fix/clean implementation of OBC with biharmonic-viscosity
498        (and also with Leith/Smagorinsky harmonic-viscosity).
499       Update output of experiment dome and exp4 (+nlfs).
500    o exp4 test-exp:
501      - turn on: nonHydrostatic, exactConserv, viscA4 (with momDissip_In_AB=F)
502        and prescribe time-varying OB (without pkg/exf) for zonal-velocity
503        at Western & Eastern Boundaries (+/- 1 per-cent); Also prescribe
504        time-varying OB at Western-side for wVel (NonHyd); update output.
505      - add secondary set-up (.nlfs) using NonLin FreeSurf (z*) + mom_vecinv
506        and biharmonic-viscosity ; also use prescribed time varying OB.
507    o pkg/obcs:
508      fix bug in OBWeta,OBWeta indices in obcs_apply_r_star.F
509    o model/src:
510      - fix startTime setting when baseTime<>0 & nIter0=0 are specified ;
511      - initialise startTime & nIter0 with unlikely values: allows to set one
512        from the other without ambiguity.
513    o seaice_obcs test-exp.:
514      add small diffK4T,S to test biharmonic diffusivity with OBCS ; update output.
515    o pkg/obcs:
516      - remove calls to OBCS_COPY_TRACER after EXCH for T,S & pTracer:
517        no longer needed with maskInC,W,S in pkg/generic_advdiff.
518    o pkg/seaice:
519      - retired SEAICE_SALINITY and SEAICE_salinity in favor of
520        SEAICE_VARIABLE_SALINITY and SIsalFRAC, respectively.
521      - fixed bug in the initialization of sea ice salinity variable HSALT
522        in seaice_init_varia.F  (removed dependence on ICE2WATR)
523    o fizhi-cs test exp.:
524       switch reference output convention for these 2 non-standard test-exp,
525        from: pgi  6.1-5 on faulks (FC.6, -of linux_ia32_pgf77+authors_fc5)
526          to: ifort 11.1 on baudelaire (FC.13, -of linux_amd64_ifort11)
527       reason: faulks might get retired soon and pkg/fizhi has still Pb with gfortran
528    o model/inc:
529      - change type of PmEpR array (from _RS to _RL) to fix restart (with "-use_r4")
530        of exp. global_ocean.90x40x15 (Suncronous time-stepping + Real-FW + NonLin-FS)
531    o pkg/seaice:
532      - fix calculation of saltFlux and saltPlumeFlux when using SEAICE_SALINITY
533    o verification/lab_sea, 1D_ocean_ice_column, seaice_obcs, global_ocean.cs32x15
534      - updated output*.txt files due to bug fixes in salt, saltPlumeFlux,
535        and initial HSALT
536    
537    checkpoint62w (2011/04/26)
538    o pkg/mom_fluxform:
539      - fix calculation of del2u & del2v (no-slip BC) when using Non-Lin Free-Surf.
540      - use parameter "sideDragFactor" in no-slip BC side-drag part.
541      - update output of exp. global_ocean.90x40x15
542    o pkg/mom_vecinv:
543      mask del2u & del2v (no effect on results, but cleaner & closer to mom_fluxform)
544    o update seaice exp.:
545      use more consistent parameter values (remove "backward compatability only":
546       ICE2WATR, SEAICE_lhSublim, SEAICE_cpAir, SEAICE_rhoAir + set rhoConstFresh)
547      related to thermodynamics seaice. Update output (fwd+AD) of exp.:
548       1D_ocean_ice_column (x2), lab_sea (x8), ioffline_exf_seaice (x2) & seaice_obcs
549    o update test-exp global_ocean.90x40x15:
550      o use default CD_CODE_OPTIONS file (Adams-Bashforth on CD Mom & Coriolis).
551      o increase Overlap size from 2 to 3 (to allow to test bi-harmonic Visc.)
552      o new pickup-files from a 100.yr run with pkg/down_slope (input.dwnslp params)
553      o turn on zStar & viscA4 in standard set-up (dir input/)
554        and update output files: output.txt & output.dwnslp.txt
555    o new AD-test experiment: obcs_ctrl
556      (From Matt MITgcm_Contrib/SOSE/BoxAdj) to test OBCS as control vector.
557    o pkg/dic:
558      fix two bugs in handling water vapour pressure in carbonate chemistry
559      and air-sea fluxes. Found by Val Bennington and Galen McKinley
560      (add DIC Option = WATERVAP_BUG to recover old results)
561      update output of tutorial_global_oce_biogeo (fwd+AD) & tutorial_dic_adjoffline
562    o adding allowInteriorFreezing option to check for water below freezing point
563      at depth and bring the negative heat anomaly to the surface level
564    o pkgs _fields_load:
565      - extend GET_PERIODIC_INTERVAL to deal with non-periodic case (from pkg/rbcs)
566      - use new S/R GET_PERIODIC_INTERVAL in pkgs:
567        bulk_force, cheapaml, cfc, dic, obcs (+ ctrl/ctrl_get_gen_rec), offline and
568        rbcs to get interpolation weights and time reccord number.
569      - fix multi-threading in aim_fields_load.F
570      - apply EXCH to temp arrays [0,1] after reading in dic_fields_load.F
571    o external_fields_load:
572      new S/R GET_PERIODIC_INTERVAL to calculate which time-record (and linear
573       interpolation weights) to use in a periodic time-series ; do not divide by
574       time-step to convert to integer:
575       - this fixes case where externForcingPeriod is not multiple of time-step
576          (Thanks to Christopher L. Wolfe)
577       - this fixes case where externForcingPeriod is an odd multiple of time-step
578         (was wrong by half of time-step); update results of tutorial_global_oce_in_p.
579      store index of time-record which is currently loaded:
580       - works without any assumption on sequence of calls to EXTERNAL_FIELDS_LOAD ;
581       - needed for the adjoint but not yet used because of extensive recomputations;
582         (to be fixed soon)
583    
584    checkpoint62v (2011/04/02)
585    o verification/1D_ocean_ice_column added to test Ian Fenty's
586      pkg/seaice_growth.F modifications for improving sea ice code differentiability
587    o update output of exp4, internal_wave, seaice_obcs, tutorial_plume_on_slope
588      (new velocity statistics after changing maskInW & maskInS definition)
589    o pkg/generic_advdiff with OBCS:
590      use maksInW & maskInS in advective flux and in biharmonic diffusion calculation:
591      equivalent to assumption of no tracer gradient outside the OB region.
592      This will affect the results if using obcs+biharmonic diffusion.
593      update output_adm.txt of experiment bottom_ctrl_5x5 (not expected since no obc)
594    o pkg/obcs:
595      change definition of maskInW & maskInS: leave OB edges inside (maskIn=1)
596      (e.g., maskInW(OB_Ie)=1 ) so that velocity normal-component at OB is still
597      in Interior region. No impact on solution but affects Velocity monitor stats.
598    o pkg/ptracers (adjoint)
599      New option AUTODIFF_PTRACERS_SPLIT_FILES for writing separate
600      tape file for each tracer
601    o model/inc/src and pkg/autodiff: add parameters that allow to turn
602      of parts of the adjoint code at run time. For now, there are only
603      useKPPinAdMode, useGMrediInAdMode, and useSEAICEinAdMode. These
604      parameters are set in data.pkg.
605      This is the lazy coward's version of doing this, but still useful for
606      debugging. In order to do it right, everything needs to be moved to
607      pkg/autodiff.
608    o pkg/autodiff: fix global_sum_tile_ad.F by resetting adsumPhi at
609      the end; this fixes the adjoint balancing code for open boundaries.
610    o pkg/ctrl:
611      - change behavior of xx_${varname}period to be analogous to exf-conventions:
612        xx_${varname}period =   0.,  means time-independent control part
613        xx_${varname}period = -12., means one-year-periodic monthly control part
614      - introduce helper routine ctrl_init_rec (part of ctrl_init.F) to
615        clean up ctrl_init a little, remove unused variables
616      - ctrl_init: fixed the handling of the start/endrec for obcs-variables
617      - move "call cal_GetMonthsRec" from s/r ctrl_get_gen to s/r ctrl_get_gen_rec
618    
619    checkpoint62u (2011/03/15)
620    o model/src:
621      - for the adjoint, change NonLin-FreeSurf initialisation sequence: fill-in
622        h0Fac from INITIALISE_FIXED and reset hFac to h0Fac in INITIALISE_VARIA ;
623      - in initialise_varia.F (except for ALLOW_DEPTH_CONTROL) remove 2nd call to
624        INI_DEPTHS, INI_MASKS_ETC & INI_LINEAR_PHISURF.
625    o pkg/ctrl/obcs/ecco:
626      - remove obsolete, unused, and partially broken code:
627        ctrl_obcsvol/ctrl_obcsbal/ctrl_volflux and code within
628        ALLOW_CTRL_OBCS_BALANCE
629      - put a stop statement into ecco/cost_obcsvol.F, but also add
630        cost_obcsvol to ecco_ad_diff.list
631    o pkg/ecco: fix counting cost function contributions for obcs (code
632      by Matt M.) and diffkr for diagnostics
633    o pkg/ecco, pkg/ctrl, pkg/grdchk:
634      add cost penalty terms for initial uvel, vvel, eta controls
635      and I/O for prior error files
636    o pkg/mdsio:
637       move mds_check4file.F from eesupp/src to mdsio_check4file.F ; add option
638       (new argument) to check for tiled file in mdsioLocalDir : this fix the
639       restart from pickup_som , pickup_dic (& dic_co2atm) with mdsioLocalDir.
640    o pkg/ctrl:
641     - ctrl_obcsbal.F, ctrl_obcsvol.F: replace last calls of ctrl_getrec by
642       ctrl_get_gen_rec, put a stop statement into ctrl_getrec. Remove
643       unused and superfluous variables
644     - ctrl_get_gen_rec.F: catch the case of xx_genperiod=0.
645    o pkg/ecco
646     - enable adjoint of Stevens OB
647    o pkg/ctrl/ecco/grdchk:
648      - remove global declaration of tmpfldx/yz and tmpfldx/yz2.
649        Fortunately, we do not have any tests for the numerous cpp-flagged
650        option of "obcs as control parameters" so we will never know
651        if this is an improvement (but at least now things compile).
652      - remove an obsolete initialization from ecco_cost_init_barfiles.F
653      - fix some problems within never used ifdefs to make the code compile
654    o pkg/obcs:
655     - Refine Interior mask definition to work with more complex OB (e.g.
656       if OB_Ie < OB_Iw); In case default setting is not sufficient, allows to
657       read in Interior mask from file. Check if Interior mask is consistent
658       with OB list of indices.
659    o pkg/seaice
660     - beginning code for a seaice tracer package,
661       but currently only used for seaice age: SEAICE_AGE
662       SEAICE_num = 4
663       iTracer = 1: area-weighted ice age
664               = 2: volume-weighted ice age
665               = 3, 4: template for one-year vs. multi-year ice
666       data.seaice has new namelist SEAICE_PARM03 for arrays related to
667       SEAICE_num
668    
669    checkpoint62t (2011/03/05)
670    o pkg/obcs:
671     - obcs_apply_uv.F: change to a 2 steps method, first tangential component;
672       then normal component. This ensures that the normal component is set
673       correctly even when it conficts with tangential setting from an other OB.
674       Affects results of test-experiment seaice_obcs: update the output.txt
675    o pkg/obcs:
676     - add code to monitor statistics of dynamical fields at OB.
677     - take OBCS_BALANCE code out of obcs_calc.F into new S/R OBCS_BALANCE_FLOW :
678       - fix to OBCS_BALANCE code (right hFac ; deal with simple OB intersection).
679       - change default: balance the Net inflow through all OB (instead of balancing
680         each OB separately); keep the old method if setting OBC_balanceFacX < 0.
681     - switch on useOBCSbalance in dome experiment: only correct Western OB outflow
682       to match Nortern OB inflow (OBCS_balanceFacN=0).
683    o pkg/seaice:
684     - redefined actual sea ice age to be ICEAGE/HEFF ifdef SEAICE_AGE_VOL
685    o pkg/seaice (seaice_growth v 1.113):
686     - reverting seaice_growth to v 1.111 in order to have code with exact bit
687       compatibility with MITgcm_contrib/ifenty/Fenty_seaice_thermo_code_merge
688       CPP option SEAICE_OCN_MELT_ACT_ON_AREA removed in v 1.112 is reintroduced
689       runtime flag SEAICEareaFormula is kept but is inactive
690    o model/src:
691     - add new param to scale curvilinear grid distance and area when using
692       a different sphere radius from the grid-files.
693     - test/check multi-threads EXCH applied to local variable (cg2d_x)
694    o adjoint:
695     - enable adjoint of Stevens OB
696     - add dummy_in_dynamics to enable extraction of ADJrhoinsitu
697       (a bit delicate because of potentially modified adj. common block structure;
698       current switch is ALLOW_AUTODIFF_MONITOR_DIAG, could be changed later)
699    
700    checkpoint62s (2011/02/15)
701    o pkg/seaice (seaice_growth v 1.112):
702     - introduce SEAICEareaFormula run time parameter, which selects the recipe for
703       computing the ice cover tendency from effective thickness tendencies. Include
704       Ian Fenty's formula.
705     - remove 'undef SEAICE_OCN_MELT_ACT_ON_AREA' which may now be achieved using
706       SEAICEareaFormula.
707     - simplify notations for the various d_HEFF/d_AREA terms.
708    o pkg/gmredi:
709     - add Boundary-Value-Problem bolus transport (Ferrari etal, OM 2010) code
710       from Christopher Wolfe (UCSD); add partial cell factor in vertical derivative;
711       add a verification test-run (front_relax.bvp) to test GM_useBVP.
712    o pkg/seaice (seaice_growth v 1.111):
713     - for SEAICE_DO_OPEN_WATER_GROWTH, replace Gael's implementation with Ian's.
714     - the new SEAICE_DO_OPEN_WATER_MELT allows open-water air-sea heat fluxes to
715       also melt ice.
716    o pkg/seaice:
717     - McPhee formula for Ice-Ocean fluxes (provided by Ian Fenty).
718     - can be used by #define MCPHEE_OCEAN_ICE_HEAT_FLUX in SEAICE_OPTIONS.h
719    o pkg/cheapaml:
720     - from Nico Wienders: New version of this package.
721    o pkg/flt:
722     - add CPP options to prevent floats to re-enter the opposite side of
723       a periodic domain (default: keep periodicity)
724    o pkg/gmredi:
725     - apply exchange to GM_PsiX,GM_PsiY to get valid values near CS-corners
726       (needed for multi-dim advection).
727     - switch on GM Advective form in global_ocean.cs32x15 and update output.txt
728    o model/src:
729     - fix wrong re-setting of addMass array (ALLOW_ADDFLUID) for atmospheric run.
730    o tutorial_tracer_adjsens
731     - extend to NLFS and replace maskC by hFacC
732       (change reference output accordingly)
733    
734    checkpoint62r (2011/01/24)
735    o testreport:
736     - add option -deloutp (-do) to remove output file after successful run
737     - add option -mfile (-mf) for MPI machile-file and for each test-exp,
738       extract exact list of procs from this file ; adjust COMMAND by
739       replacing TR_MFILE & TR_NPROC with local machile-file & Nb of MPI procs.
740  o pkg/ecco: replace old ineffective way of penalizing deviations from  o pkg/ecco: replace old ineffective way of penalizing deviations from
741    first guess (of obcs) with code by Matt Mazloff    first guess (of obcs) with code by Matt Mazloff
742  o pkg/ctrl: add code that allows the use of ALLOW_OBCS?_CONTROL without  o pkg/ctrl: add code that allows the use of ALLOW_OBCS?_CONTROL without
743    exf/cal    exf/cal
744    - ctrl_init: fix initialisation (computation of diffrec) for obcs-ctrl    - ctrl_init: fix initialisation (computation of diffrec) for obcs-ctrl
745      and handle case of obcs?period == 0      and handle case of obcs?period == 0
746      Not affected by these changes: There is still an inconsistency in      Not affected by these changes: There is still an inconsistency in
747      the computation of start/endrec for the obcs variables for ALLOW_CAL      the computation of start/endrec for the obcs variables for ALLOW_CAL
748      defined (see comment by Menemenlis, 7-Mar-2003)      defined (see comment by Menemenlis, 7-Mar-2003)
749    - make S/R ctrl_get_gen_rec return something meaningful even if    - make S/R ctrl_get_gen_rec return something meaningful even if
750      ALLOW_CAL is not defined      ALLOW_CAL is not defined
751    - ctrl_getobcs?: simplify code so that it does not need to use    - ctrl_getobcs?: simplify code so that it does not need to use
752                     S/R exf_swapfields                     S/R exf_swapfields
753  o testreport:  o testreport:
754     allow to test MPI with different number of processors for each experiment:     allow to test MPI with different number of processors for each experiment:
# Line 24  o testreport: Line 757  o testreport:
757  o model/src, pkg/autodiff, pkg/ecco, pkg/seaice:  o model/src, pkg/autodiff, pkg/ecco, pkg/seaice:
758    alleviate the need for additional 3D tapes when using    alleviate the need for additional 3D tapes when using
759    the non-inear free surface in the adjoint. This is done    the non-inear free surface in the adjoint. This is done
760    by adding an 'update' of hfacc etc. to their current    by adding an 'update' of hfacc etc. to their current
761    value at the beginning of forward_step.F.    value at the beginning of forward_step.F.
762  o pkg/seaice  o pkg/seaice
763    seaice_lsr: move stress-divergence computations out of outer pseudo-    seaice_lsr: move stress-divergence computations out of outer pseudo-
764    timestepping loop    timestepping loop
Legend:
Removed from v.1.1367  
changed lines
  Added in v.1.1508
  ViewVC Help
Powered by ViewVC 1.1.22