/[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 1.1779 by jmc, Sun Dec 9 21:01:16 2012 UTC revision 1.1802 by jmc, Sun Jan 6 20:36:27 2013 UTC
# Line 4  $Name$ Line 4  $Name$
4      Notes on tags used in MITgcmUV      Notes on tags used in MITgcmUV
5      ==============================      ==============================
6    
7    o verification/offline_exf_seaice:
8      - dyn_lsr  : use pkg/seaice S/R (instead of pkg/thsice) to advect sea-ice;
9      - dyn_jfnk : use pkg/thsice S/R (instead of pkg/seaice) to advect sea-ice.
10    o model/src:
11      - implement Finite-Volume method for (hydrostatic) presure gradient
12        from S.-J. Lin (QJRMS 1997), for atmosphere using sigma-coordinate.
13    o pkg/seaice: implement EVP* by Lemieux et al. (2012).
14      - In order to be compatible with the default EVP code, this requires changing
15        the EVP code around a little; and because any small change immediately
16        breaks the lab_sea.hb87 verification experiment, the reference output is
17        updated two times (each time the remaining aggrement was only 2 digits):
18        - after replacing 1/dt by 1*recip_dt with a precomputed recip_dt
19        - after replacing 1/(1+0.5*dt/tau) by a/(a+1), with a=2*tau/dt
20        The adjoint test lab_sea.evp also needed to be updated.
21      - Turn on EVP* by setting SEAICE_deltaTevp as before, and setting
22        SEAICEuseEVPstar=.true.; for EVP* SEAICE_deltaTevp does not need to be
23        a factor of SEAICE_deltaTdyn, and you can set the number of EVP steps
24        explicitly with SEAICEnEVPstarSteps (thereby overriding the default of
25        INT(SEAICE_deltaTdyn/SEAICE_deltaTevp)). Controlling EVP* is deliberately
26        kept obscure, because I need to test it a little more.
27    o pkg/cheapaml:
28      - fix few incorrect type (_RL/_RS) in subroutine calls.
29    o pkg/seaice:
30      - new S/R SEAICE_MAP_RS2VEC (similar to SEAICE_MAP2VEC) to map _RS arrays
31        to _RL vector; this fixes JFNK code when compiling with "-use_r4" option.
32    o model/src:
33      - add a warning if using select_rStar=2 with nonlinFreeSurf <> 4;
34        change (in input_ad*/data) select_rStar from 2 to 1 for AD verification
35        experiments global_ocean.cs32x15 & tutorial_tracer_adjsens (+ update output)
36    o pkg/cheapaml:
37      - replace "273.1 _d 0" by celsius2K in pkg/cheapaml/cheapaml.F and
38        update output of experiment cheapAML_box.
39      - store in CHEAPAML.h coeffs ssq[0:2] for saturation specific humidity,
40        and set them only in one place (cheapaml_readparms.F).
41      - replace "273.16 _d 0" by celsius2K & replace "3.14159 _d 0" by constant "PI"
42      - remove re-setting of main-model parameter "celsius2K" (in cheapaml_readparms)
43        and re-setting of cheapAML params "lath" & "p0" (in cheapaml_init_varia).
44    o model/src/the_main_loop.F:
45      - fix revision 129 : need both ALLOW_ECCO and ALLOW_ECCO_EVOLUTION.
46    o pkg/seaice:
47      - remove legacy branch code.
48      - retire SEAICE_GROWTH_LEGACY and SEAICE_CAP_HEFF accordingly.
49    o verification/*/code*/SEAICE_OPTIONS.h:
50      - remove documentation for retired CPP flags (SEAICE_GROWTH_LEGACY,
51        SEAICE_CAP_HEFF).
52    o pkg/cheapaml:
53      - use standard routine (S/R GET_PERIODIC_INTERVAL) to decide when it is time
54        to load new forcing fields.
55      - add missing "_d 0" where it's needed; + some editing (improve indentation).
56      - remove switch to "-r8" compiler option (no longer needed) in verification
57        cheapAML_box, keep same output (small trucation error differences).
58    o verification/offline_exf_seaice:
59      - add thermodynamics only test.
60    o verification/cheapAML_box:
61      - split the domain in 8 (2x4) tiles; add MPI and multi-threaded test;
62        keep same output (same 14 digits for cg2d).
63      - switch on "-r8" compiler option (some numerical values are missing
64        the "_d 0") and update output (only 6 digits for cg2d were preserved).
65    
66    checkpoint64b (2012/12/24)
67    o pkg/cheapaml:
68      - remove un-used variables (Qnet,EmPmR) from pickup files.
69      - store tendency of advected variables into cheapaml pickup file
70        and fix the AB-2 restart.
71    o verification/offline_exf_seaice and lab_sea adjoint:
72      - switch to new code (#undef SEAICE_GROWTH_LEGACY).
73      - switch to 4 time steps in offline, consistent with lab_sea.
74      - match parameters from SEAICE_GROWTH_LEGACY branch in data.seaice.
75      - update results after change of experiment seaice.
76    o pkg/seaice:
77      - remove option SEAICE_MULTICATEGORY; has been replaced with run-time
78        parameter SEAICE_multDim. Issue a warning if SEAICE_MULTICATEGORY is
79        still defined since SEAICE_multDim needs to be set to MULTDIM value (from
80        SEAICE_SIZE.h) to recover previous #define SEAICE_MULTICATEGORY default.
81    o model/src:
82      - add run-time parameter to select CG2D solver minimum-residual solution
83        (in case of a poor convergence).
84    o pkg/seaice/seaice_lsr.F
85      - move calculation of matrix coefficients into new
86        S/R SEAICE_CALC_LSR_COEFFS; for now this routine remains in the same
87        file; the main purpose of this change is to be able to reuse the code
88        in S/R SEAICE_PRECONDITIONER
89      - modify the computation of the residual so that the mass of sea-ice
90        is not used twice; this simplifies the calculations, but since
91        the residual is never used except when LSR_mixIniGuess.ge.4---and this is
92        not tested in testreport---this does not affect the testreport results.
93    o verification/seaice_obcs:
94      - disable SEAICE_GROWTH_LEGACY; reset input files to match those of
95        verification/lab_sea/input.salt_plume; changes results/output.txt*
96    o model/src:
97      - adapt quasi-hydrostatic formulation to work with p or p* coordinate.
98    o pkg/seaice:
99      - improve description of parameters in seaice_fgmres.F and start
100        cleaning up the code
101      - change name of SCALPROD to SEAICE_SCALPROD and modify meaning of
102        second argument
103      - add a metric based on grid cell area to SEAICE_SCALPROD; update
104        results of offline_exf_seaice.dyn_jfnk
105      - compute initial residual norm in seaice_jfnk based on SEAICE_SCALPROD
106    o pkg/aim_v23:
107      - fix an other bug in dissipation heating (grid-cell area was missing).
108    o model/src + pkg/seaice:
109      - add global (multi-tile) linear solver for pair of tri-diagonal system
110        along X and Y lines, respectively for U and V component.
111        Note: 1) MPI and cube-exchange not yet coded.
112            2) probably not accurate for poorly conditioned / large size problem.
113      - add option for SEAICE_LSR to use global (multi-tile) tri-diagonal solver;
114        default: not compiled and not used.
115    o pkg/seaice:
116      - first attempt at a multi-threaded fgmres solver for the JFNK solver:
117        introduce bi/bj loops and use global_sum_tile_rl instead of
118        mpi_allreduce in seaice_fgmres.F
119      - improve diagnostics output for fgmres solver (seaice_fgmres.F)
120      - replace a few unnecessary goto statements by if-blocks  (seaice_fgmres.F)
121      - add a verification experiment for JFNK in offline_exf_seaice
122  o verification:  o verification:
123    - change (for now, only fwd test) offline_exf_seaice experiment:    - change (for now, only fwd test) offline_exf_seaice experiment:
124      use a re-entrant channel with idealized costline (trapezoidal shape)      use a re-entrant channel with idealized coastline (trapezoidal shape)
125      at southern boundary to explore seaice dynamics and thermodynamics      at southern boundary to explore seaice dynamics and thermodynamics
126      with simple forcing. Main set-up has dynamics+thermo and SST relaxation;      with simple forcing. Main set-up has dynamics+thermo and SST relaxation;
127      secondary test (input.thsice) use thermodynamics only with SST relaxation.      secondary test (input.thsice) use thermodynamics only with SST relaxation.
Legend:
Removed from v.1.1779  
changed lines
  Added in v.1.1802
  ViewVC Help
Powered by ViewVC 1.1.22