4 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
5 |
============================== |
============================== |
6 |
|
|
7 |
|
o pkg/autodiff: zero_adj resets an ajoint variable to zero (in adjoint mode) |
8 |
|
o pkg/ctrl: ctrl_bound forces parameter corrections to stay in |
9 |
|
a bounded range (in forward mode) |
10 |
|
o pkg/ecco, ctrl, autodiff and mom_common: |
11 |
|
internal parameter controls, completements of implementations |
12 |
|
o add few EXCH S/R (UV_BGRID, UV_DGRID, SM) for B & D-Grid velocity & 2nd Moment Sxy. |
13 |
|
o pkg/cd_code: use the right EXCH calls (for CS-grid); but problem remains |
14 |
|
at face corners. |
15 |
|
o pkg/generic_advdiff: |
16 |
|
- in GAD.h, define as parameters (iMinAdvR,iMaxAdvR,jMinAdvR,jMaxAdvR) |
17 |
|
loop range indices for computing vertical advection tendency |
18 |
|
and do the calculation in the interior only (no overlap) ; |
19 |
|
for now, only implemented for SOM advection (to save CPU time). |
20 |
|
- add argument "withSigns" to S/R FILL_CS_CORNER_TR_RL (needed for SOM_xy moment) |
21 |
|
o exch2 adjoint |
22 |
|
First set of changes: |
23 |
|
- add exch2 hand-written adjoint templates |
24 |
|
- exch2 hand-written no longer use TAF-flag 'nownew_arg' |
25 |
|
This will require cleanup in pkg/autodiff/*.F |
26 |
|
(changed routines currently kept separate in verif. code/) |
27 |
|
- Comment all relevant #ifndef ALLOW_AUTODIFF_TAMC |
28 |
|
that used to hide exch2 or cubed-sphere specific code |
29 |
|
(commented via 'cph-exch2') |
30 |
|
o pkg/seaice |
31 |
|
- Close southern and western boundaries for UICE/VICE when useOBCS=.TRUE. |
32 |
|
o pkg/bulk_force: fix loading into Qsw of Short-Wave flux; and update |
33 |
|
output of global_ocean.cs32x15.thsice . |
34 |
|
|
35 |
|
checkpoint59e |
36 |
|
o exch2: new S/R that only use the cube-tracer (center-position) exchange: |
37 |
|
for B-grid vector (not yet tested) and C-grid vector (<- only used |
38 |
|
when option W2_USE_R1_ONLY in W2_OPTIONS.h is defined). |
39 |
|
o starting implementation of Duffy et al. (GRL 1999) salt plume scheme |
40 |
|
(#ifdef ALLOW_SALT_PLUME) |
41 |
|
o gather_2d/scatter_2d: removed halo regions |
42 |
|
o gmredi: add option (GM_taper_scheme='fm07') for Ferrari & McWilliams 2007 |
43 |
|
scheme and the corresponding test case (under front_relax dir). For now, |
44 |
|
only available with Skew-flux form and not yet tested in realistic set-up. |
45 |
|
adjoint: store directives updated, but gradient checks deteriorated, |
46 |
|
so maybe GM/Redi adjoint is now broken. |
47 |
|
o seaice adjoint |
48 |
|
* add AREA, HEFF, HSNOW as control variable |
49 |
|
(should not change lab_sea results) |
50 |
|
* CTRL_OPTIONS.h needs new flag: CTRL_SET_OLD_MAXCVARS_40 |
51 |
|
o pkg/seaice: |
52 |
|
* added computation of saltFlux in seaice_growth; |
53 |
|
For time being, sea ice salinity is assumed constant |
54 |
|
contribution of snow flooding to freshening of sea ice is neglected. |
55 |
|
SEAICE_salinity is set in SEAICE_PARM01 of data.seaice and it |
56 |
|
defaults to 0.0 in order to maintain backward compatibility. |
57 |
|
Typical values for SEAICE_salinity are 4 to 10 g/kg. |
58 |
|
* replaced dRf(1) with dRf(kSurface)*hFacC(kSurface) in seaice_growth.F |
59 |
|
* added SEAICE_availHeatFrac, which specifies fraction of surface level |
60 |
|
heat content used to melt or grow ice |
61 |
|
* commented out cap of HICE: it helps keep ice thinner |
62 |
|
* reintroduced cap of HEFF via MAX_HEFF and SEAICE_CAP_HEFF option |
63 |
|
* applied OBCS T/S mask to HEFFM sea ice mask |
64 |
|
|
65 |
|
checkpoint59d |
66 |
|
o standard name: {PKG}_DIAGNOSTICS_STATE to fill diagnostics array |
67 |
|
for {PKG} state variables, and called from DO_STATEVARS_DIAGS. |
68 |
|
(keep {PKG}_DIAGNOSTICS_FILL for non-state variables diagnostics) |
69 |
|
o pkg/gmredi (gmredi_readparms.F): add some missing "_d 0" |
70 |
|
and change default value of parameter "GM_Small_Number"; |
71 |
|
update output of exp: lab_sea (x2), cfc_example, global_ocean.90x40x15, |
72 |
|
global_ocean.cs32x15 (x4), global_ocean_ebm, global_with_exf, |
73 |
|
tutorial_global_oce_biogeo, tutorial_global_oce_latlon |
74 |
|
o pkg/ecco: cost function when applying pkg/smooth to controls |
75 |
|
o pkg/ctrl: add forcingPrecond run time parameter |
76 |
|
o pkg/seaice: |
77 |
|
- replace ZMIN=4.e8 with ZMIN=0. in dynsolver.F as in C-grid code |
78 |
|
- fix bug in metric terms of u-equation in LSR solver (implicit part), |
79 |
|
changes verification/lab_sea |
80 |
|
- rearrange implicit part of u-equation in LSR solver to make it easier to |
81 |
|
read, understand, and debug. This causes changes at the truncation level. |
82 |
|
- fix (serious) bug in the boundary conditions (sign error in the implicit |
83 |
|
part) of the LSR solver. Changes all LSR results. |
84 |
|
- fixed problem with HSNOW initialization to avoid thick snow when AREA |
85 |
|
is initilized with small values - does not change lab_sea results. |
86 |
|
- added HsnowFile initialization capability |
87 |
|
o pkg/smooth: addition to model/src and pkg/ctrl so allow |
88 |
|
initialization of pkg/smooth and application to control vector |
89 |
|
o pkg/seaice: fix another bug in the LSR no-slip boundary conditons |
90 |
|
o dependencies: pkg/profiles requires pkg/cal |
91 |
|
o pkg/profiles: |
92 |
|
- added option ALLOW_PROFILES_GENERICGRID |
93 |
|
-> pkg/profiles can thus be used with any grid (e.g. cs32x15) assuming |
94 |
|
that interpolation information (grid points and coefficients) is provided |
95 |
|
within the netcdf input files. |
96 |
|
-> sample matlab scripts can be found in MITgcm_contrib/gael |
97 |
|
that allow you to prepare netcdf input files. |
98 |
|
- also modified: cleaner include statements, cleaner error print statements, |
99 |
|
cleaner stops when error, and more tests of the netcdf input files consistency. |
100 |
|
o additional test (in global_ocean.cs32x15) using seaice+thsice pkgs. |
101 |
|
o pkg/seaice: |
102 |
|
- move seaice-diagnostics from seaice_do_diags to do_statevars_diags; this |
103 |
|
requires a new routine: seaice_diagnostics_fill.F |
104 |
|
- make sure that seaice does not move if not forced (pressure replacement |
105 |
|
method, Hibler and Ip, 1995), changes results |
106 |
|
- and new damping criterion for EVP solver according to Hunke, JCP, 2001 |
107 |
|
- SEAICE_evpDampC = > 0. turns it on (615. is the recommended value). |
108 |
|
It is turned off by default (=-1.), because it changes the results. |
109 |
|
- make it possible to read AREA from a file, too. |
110 |
|
|
111 |
|
checkpoint59c |
112 |
|
o add S/R and diagnostics for oceanic mixed-layer depth |
113 |
|
o move GMRedi call after all vertical mixing schemes (including KPP) |
114 |
|
o pkg/thsice: |
115 |
|
- ensure iceMask > iceMaskMin : this problem appeared after May-04 modifs |
116 |
|
(melt only laterally if thin ice): update output of global_ocean.cs32x15.thsice |
117 |
|
o pkg/obcs: |
118 |
|
- bug fix for balance code |
119 |
o pkg/seaice: |
o pkg/seaice: |
120 |
- allow values of SEAICE_elaticParm > 1 |
- allow values of SEAICE_elasticParm > 1 |
121 |
- make lsr boundary conditions free slip (by default) |
- add new parameter SEAICE_evpTauRelax as an alternative |
122 |
|
to SEAICE_elasticParm |
123 |
|
- make lsr boundary conditions free slip (by default), changes result |
124 |
- add code for no slip boundary conditions for lsr |
- add code for no slip boundary conditions for lsr |
125 |
- set minimum of zeta to zero (and not 4e8) (changes results) |
- set minimum of zeta to zero (and not 4e8) (changes results) |
126 |
|
- add a cap on AREA after advecting AREA if seaice_growth is not called |
127 |
o change calculation of Visbeck K: |
o change calculation of Visbeck K: |
128 |
- no longer depend on tapering scheme; instead, compute slope (limited |
- no longer depend on tapering scheme; instead, compute slope (limited |
129 |
by GM_maxSlope). this is equivalent to what was comuted before with |
by GM_maxSlope). this is equivalent to what was comuted before with |
130 |
clipping or linear tappering. |
clipping or linear tappering. |
131 |
- fix vertical averaging (wrong by 1 level) |
- fix vertical averaging (wrong by 1/2 level) |
132 |
- leave the old version (but will not be maintained) within |
- leave the old version (but will not be maintained) within |
133 |
#ifdef OLD_VISBECK_CALC / #endif |
#ifdef OLD_VISBECK_CALC / #endif |
134 |
|
- update output of exp. ideal_2D_oce |
135 |
o change error catch in calc_r_star to improve vectorization |
o change error catch in calc_r_star to improve vectorization |
136 |
o exf runoff and seaice: remove ALLOW_SEAICE from exf pkg files and |
o exf runoff and seaice: remove ALLOW_SEAICE from exf pkg files and |
137 |
add #define ALLOW_RUNOFF in SEAICE_OPTIONS.h (when not using exf). |
add #define ALLOW_RUNOFF in SEAICE_OPTIONS.h (when not using exf). |
138 |
|
o adjoint: |
139 |
|
Re-organized adjoint checkpointing according to Matt Mazloff |
140 |
|
(but with modifs/completions/cleanups) |
141 |
|
Storing is now bundled into large arrays to optimized I/O |
142 |
|
(and second step in cleanup of the_main_loop) |
143 |
|
|
144 |
checkpoint59b |
checkpoint59b |
145 |
o coupler (pkg/atm_ocn_coupler, pkg/ocn_compon_interf, pkg/atm2d): |
o coupler (pkg/atm_ocn_coupler, pkg/ocn_compon_interf, pkg/atm2d): |
193 |
o pkg/thsice: |
o pkg/thsice: |
194 |
fix few little problems, in thsice_calc_thickn.F: |
fix few little problems, in thsice_calc_thickn.F: |
195 |
- growth vertically (and not laterally) if iceFrac == iceMaskMax |
- growth vertically (and not laterally) if iceFrac == iceMaskMax |
196 |
- melt laterally only if hIce < hThinIce (as the comments say) |
- melt only laterally if hIce < hThinIce (as the comments say) |
197 |
and in thsice_extend.F: |
and in thsice_extend.F: |
198 |
- allow to form ice even when iceFrac == iceMaskMax (by increasing thickness) |
- allow to form ice even when iceFrac == iceMaskMax (by increasing thickness) |
199 |
- start to form ice as soon as the minimum ice-volume is reachable. |
- start to form ice as soon as the minimum ice-volume is reachable. |