4 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
5 |
============================== |
============================== |
6 |
|
|
7 |
|
checkpoint61d |
8 |
|
o adjoint modifs |
9 |
|
- adapt to new field rhoInSitu |
10 |
|
- allow to revert from StoreDynVars2D, StoreDynVars3D |
11 |
|
to individual fields |
12 |
|
(switches AUTODIFF_USE_OLDSTORE_2D, AUTODIFF_USE_OLDSTORE_3D) |
13 |
|
o store 3-D (in-situ) density in commom block (DYNVARS.h) |
14 |
|
- save 1 rho computation. |
15 |
|
- no longer modify T & S inland when using down_slope pkg. |
16 |
|
o gad_som : allow to switch to SOM advection for T & S without corresponding |
17 |
|
pickup files (using new S/R: MDS_CHECK4FILE). |
18 |
|
o keep removing _BEGIN/_END_MASTER between MDSIO S/R call |
19 |
|
(for useSingleCpuIO to work in multi-threaded) |
20 |
|
o pkg/shelfice: |
21 |
|
- disentangle parameters: retire shelfIceFile in data/PARM05 and substitute |
22 |
|
with SHELFICEtopoFile in data.shelfice |
23 |
|
- move code for reading ice shelf topography and modifying hFacC into separate |
24 |
|
subroutines of the shelfice package |
25 |
|
- separate initialisation of "fixed" and "variable" variables |
26 |
|
- kTopC is not (yet) replaced by kSurfC, although this is possible |
27 |
|
o fix broken diagnostic SIatmFW and changed the units from m/s to kg/m^2/s |
28 |
|
o add initialisation of deepFac if using Pcoords (new S/R set_grid_factors) |
29 |
|
and rename: |
30 |
|
ini_phiref.F --> set_ref_state.F (+ set anelastic factors) |
31 |
|
ini_reference_state.F --> load_ref_files.F (- set anelastic factors) |
32 |
|
o change FIND_RHO_SCALAR : return rho (instead of rho - rhoConst) |
33 |
|
|
34 |
|
checkpoint61c |
35 |
|
o add option ALLOW_ADDFLUID and run-time selector "selectAddFluid" |
36 |
|
to add mass source/sink of fluid in the interior (3-D generalisation |
37 |
|
of oceanic real-fresh water flux). |
38 |
|
o pkg/monitor: fix recip_dr argument in mon_advcflw2 call (-> advcfl_W_hf_max) |
39 |
|
o vermix test experiment: switch from LINEAR EOS to MDJWF (not tested before) |
40 |
|
o new pkg: down_slope |
41 |
|
density-driven downsloping flow parameterisation (Tellus, 1999, 412-430). |
42 |
|
- presently, modify T & S inland (dwnslp_calc_rho.F). |
43 |
|
can be avoided if rho was stored in common bloc and used in calc_phi_hyd. |
44 |
|
- implementation for ocean in P-coords not tested. |
45 |
|
- add secondary test to global_ocean.90x40x15 (input.dwnslp) to check this pkg. |
46 |
|
o move kSurfC,kSurfW & kSurfS from SURFACE.h to GRID.h (next to kLowC) |
47 |
|
(to facilitate conversion of oceanic pkg to p-coordinate). |
48 |
|
o change FIND_RHO to FIND_RHO_2D : |
49 |
|
pass arguments tFld & sFld as local (tile) 2D-arrays ; +fix EOS_CHECK. |
50 |
|
o new pkg: cheapaml |
51 |
|
|
52 |
|
checkpoint61b |
53 |
|
o exch2: |
54 |
|
change index bounds in rx2_cube exchanges (new S/R: EXCH2_GET_UV_BOUNDS) |
55 |
|
- no longer depend on the order sequence (N,S,E,W). |
56 |
|
- 3rd exchange no longer needed (tested with 24 tiles). |
57 |
|
- same modif to hand-written adjoint S/R (global_ocean.cs32x15: zero diff) |
58 |
|
- exch_UV_A-grid readily available (but not yet tested). |
59 |
|
comment out 3rd exch call and remove copy of V to U at SE & NE face-corners |
60 |
|
(both are no longer needed) in exch_uv: no effect in forward run; but |
61 |
|
get truncation error differences with adjoint exch. |
62 |
|
o exch2: |
63 |
|
- change index-bounds storage (moved from target to local tile, |
64 |
|
more intuitive this way) ; rename/remove some variables. |
65 |
|
- change utils/exch2/matlab-topology-generator & code_mods accordingly. |
66 |
|
|
67 |
|
checkpoint61a |
68 |
|
o exch2/matlab-topology-generator: |
69 |
|
fix bug in index bounds for N-S internal exchange (hard to tell if this |
70 |
|
had any effect on the solution). |
71 |
|
o pkg/mnc: small fix in mnc_handle_err: make sure that length of msgbuf |
72 |
|
is exceeded |
73 |
|
o testreport: modify syntax of awk, so that it works also with |
74 |
|
awk, and not only with nawk and gawk |
75 |
|
|
76 |
|
checkpoint61 |
77 |
|
o autodiff |
78 |
|
Change default for TAF usage by removing argument '-nonew_arg' |
79 |
|
This changes TAF subroutine argument list (ordering of active var.) |
80 |
|
Omission of this flag leads to incompatibility w.r.t. TAMC |
81 |
|
To restore TAMC-compatibility, need following modifs: |
82 |
|
* use adoptfile tools/adjoint_options/adjoint_tamc_compatibility |
83 |
|
* use CPP option #define AUTODIFF_TAMC_COMPATIBILITY |
84 |
|
Tested TAF version is 1.9.22 |
85 |
|
N.B.: exch2 hand-written adjoint code currently not TAMC compatible |
86 |
|
|
87 |
|
checkpoint60 |
88 |
|
--->>> This is the last checkpoint for which generating the adjoint |
89 |
|
--->>> is done via the TAMC-compatibility TAF-flag "-nonewarg" |
90 |
|
--->>> by default. |
91 |
|
--->>> Future checkpoints will ommit this flag by default. |
92 |
|
o pkg/generic_advdiff, os7mp: simplify and avoid division by zero. |
93 |
|
o pkg/grdchk: Add gradient check to etan0 perturbations. |
94 |
|
o pkg/exf: read ice area from file (#ifdef ALLOW_ICE_AREAMASK). |
95 |
|
o pkg/autodiff: more flexibility in zeroadj. |
96 |
|
model/src: modifying do_oceanic_phys.F accordingly. |
97 |
|
o pkg/ctrl: Bounds for (kapgm,kapredi,eddypsi) parameter estimation. |
98 |
|
o model/src and pkg/gmredi: GMREDI_WITH_STABLE_ADJOINT CPP option |
99 |
|
special setup of gmredi for which the adjoint is stable |
100 |
|
enough for parameter optimization. |
101 |
|
o bridging the gap between eddy stress and GM. |
102 |
|
-> eddyTau is replaced with eddyPsi (eddyTau = f x rho0 x eddyPsi) |
103 |
|
along with a change in CPP option (now ALLOW_EDDYPSI). |
104 |
|
-> when using GM w/ GM_AdvForm: |
105 |
|
The total eddy streamfunction (Psi = eddyPsi + K x Slope) |
106 |
|
is applied either in the tracer Eq. or in momentum Eq. |
107 |
|
depending on data.gmredi (intro. GM_InMomAsStress). |
108 |
|
-> ALLOW_EDDYPSI_CONTROL for estimation purpose. |
109 |
|
The key modifications are in model/src/taueddy_external_forcing.F |
110 |
|
pkg/gmredi/gmredi_calc_*F pkg/gmredi/gmredi_*transport.F |
111 |
|
|
112 |
|
checkpoint59r |
113 |
|
o new S/R (pkg/mdsio/mdsio_facef_read.F) reading of sNx+1 x sNy+1 array |
114 |
|
from per-face file (take S/R READSYMTILE_RS out of ini_curvilinear_grid.F); |
115 |
|
fix option to read coriolis(@ cell-corner) from files with CS-grid. |
116 |
|
o pkg/mnc: add some attributes to standard 1D-coordinate variables |
117 |
|
o pkg/diagnostics: enforce missing_value attribute for scalar variables at |
118 |
|
mass points, when using mnc for output |
119 |
|
- add runtime parameter (missing_value, missing_value_int) to override |
120 |
|
default (which is the value of getcon('UNDEF')) |
121 |
|
o pkg/mnc: prepare mnc for the actual use of missing values with the |
122 |
|
diagnostics package: small fix in mnc_cw_readwrite.template and more |
123 |
|
initialisation of never used fields |
124 |
|
- replace mnc_common.h and mnc_id_header.h with corresponding upper case |
125 |
|
versions |
126 |
|
o pkg/thsice/thsice_ini_vars.F added default initial values so that |
127 |
|
thermodynamics sea ice solver does not blow up when some initial fields |
128 |
|
are missing. |
129 |
|
o move from external_forcing_surf.F to ini_forcing.F |
130 |
|
the p-coord. hack consisting in using pLoad to set phi0surf ; |
131 |
|
update output of tutorial_global_oce_in_p (affected @ truncation level). |
132 |
|
o Freshen up a bit hs94.1x64x5, notably: |
133 |
|
remove rotationPeriod and tRef=5*0. (adjoint only), and add |
134 |
|
exactConserv=.TRUE. |
135 |
o pkg/generic_advdiff: |
o pkg/generic_advdiff: |
136 |
change option GAD_ALLOW_SOM_ADVECT to GAD_ALLOW_TS_SOM_ADV which only |
change option GAD_ALLOW_SOM_ADVECT to GAD_ALLOW_TS_SOM_ADV which only |
137 |
applies to files where Temperature & Salinity 2nd Order moments are used. |
applies to files where Temperature & Salinity 2nd Order moments are used. |