4 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
5 |
============================== |
============================== |
6 |
|
|
7 |
|
o verification/global_ocean.cs32x15/input.seaice : |
8 |
|
- add SEAICEheatConsFix in data.seaice, ensuring closed ocean+ice heat budget. |
9 |
|
- add diagnostics (commented out) for budget test using my matlab codes (gcmfaces). |
10 |
|
- update results (output.seaice.txt) accordingly. |
11 |
|
o verification/lab_sea/input/data.seaice : |
12 |
|
- remove SEAICEheatConsFix, which is now be tested |
13 |
|
in global_ocean.cs32x15. The presently coded fix |
14 |
|
is only consistent with real fresh water + nlfs. |
15 |
|
- update results (output.txt) accordingly. |
16 |
|
o pkg/salt_plume: added diagnotic for salt tendency (oceSPtnd). |
17 |
|
o AD_CONFIG.h file: |
18 |
|
- in genmake2: remove "#undef ALLOW_ECCO_OPTIMIZATION" from AD_CONFIG.h; |
19 |
|
- remove #include "AD_CONFIG.h" from all included OPTIONS files and |
20 |
|
add it explicitly in each fortran src file where it is needed. |
21 |
|
motivation: header file AD_CONFIG.h is generated/modified after "make depend" |
22 |
|
thus introducing potentially wrong dependency; therefore, better to include |
23 |
|
it only where it's really needed, and after any other included OPTIONS file |
24 |
|
(less tempting to bring conditional includes according to ALLOW_ADJOINT_RUN |
25 |
|
or ALLOW_TANGENTLINEAR_RUN). |
26 |
|
|
27 |
|
checkpoint63p (2012/07/06) |
28 |
|
o pkg/autodiff: |
29 |
|
- rename S/R INI_AUTODIFF to AUTODIFF_INIT_VARIA ; |
30 |
|
- create new header file "AUTODIFF_PARAMS.h" for pkg/autodiff parameters |
31 |
|
which are now read from new parameter file "data.autodiff" (inAdExact |
32 |
|
previously in "data", use{KPP/GMRedi/SEAICE}inAdMode previously in data.pkg) |
33 |
|
- add file "data.autodiff" in all verification/*/input_ad/ directories. |
34 |
|
- add run-time parameter to turn On/Off SEAICEuseFREEDRIFT in adjoint/forward |
35 |
|
mode (used in global_ocean.cs32x15.seaice_dynmix test exp.) |
36 |
|
o verification/tutorial_tracer_adjsens/ |
37 |
|
- add extra sub-verification which uses "SOM" advection scheme |
38 |
|
(2nd Order-Moment Advection Scheme, Prather Limiter) with |
39 |
|
#define GAD_ALLOW_TS_SOM_ADV in GAD_OPTIONS.h |
40 |
|
and in data: |
41 |
|
multiDimAdvection=.TRUE., |
42 |
|
tempAdvScheme=81, |
43 |
|
saltAdvScheme=81, |
44 |
|
o model/src: |
45 |
|
- introduce ALLOW_BALANCE_RELAX which allow the removal |
46 |
|
of the global mean of relaxation terms by setting |
47 |
|
balanceThetaClimRelax and balanceSaltClimRelax |
48 |
|
- disable balanceEmPmR and balanceQnet in the case when useSeaice. |
49 |
|
This case is now treated appropriately in seaice_growth.F |
50 |
|
o pkg/seaice: |
51 |
|
- do balanceEmPmR and balanceQnet based on the net fluxes |
52 |
|
(SItflux) that enter the ocean+ice system rather than QNET |
53 |
|
and EMPMR. Works best in real fresh water. |
54 |
|
- SIatmFW, SIatmQnet, and SItflux computations had to be moved |
55 |
|
out of the diagnotics bloc, and stored in global arrays. |
56 |
|
|
57 |
|
checkpoint63o (2012/06/29) |
58 |
|
o vertical mixing schemes: |
59 |
|
- remove unused S/R CALC_BUOYANCY (left from branch-atmos merging) |
60 |
|
- replace computation of density (in Function STATE1, opps_calc.F) by a call |
61 |
|
to S/R FIND_RHO_SCALAR |
62 |
|
- pass sigmaR to S/R PP81_CALC, MY82_CALC (but not yet used) and to GGL90_CALC |
63 |
|
and use it for N^2 calculation in ggl90_calc.F (save 2 density computations); |
64 |
|
affects output (vermix.ggl90) at truncation level. |
65 |
|
o testreport: |
66 |
|
- change adjustment of SIZE.h for mpi+mth testing: preserve enough tiles |
67 |
|
(nSx,nSy) for the number of threads (from eedata.mth) |
68 |
|
o model/src: |
69 |
|
- remove 2nd set of calls at the top of initialise_varia.F (Adjoint part) |
70 |
|
related to Non_Lin Free-Surf or Bottom-Control (but not longer necessary). |
71 |
|
o verification/tutorial_tracer_adjsens/ |
72 |
|
- compile with ALLOW_CG2D_NSA |
73 |
|
- run with nonlinFreeSurf=3 |
74 |
|
- update reference output_adm.txt |
75 |
|
o tools/genmake2: |
76 |
|
- add "-DHAVE_NETCDF" to the DEFINES list when NetCDF is available ; |
77 |
|
refine usePROFILES switching off (only when NetCDF is not available). |
78 |
|
o verification (atmospheric set-up): |
79 |
|
- change eosType='IDEALGAS' to eosType='IDEALG' to match variable declaration |
80 |
|
o pkg/profiles : refine profiles and files handling. |
81 |
|
o model/src and pkg/ecco : introduce run-time switch usePROFILES. |
82 |
|
o model/src/the_main_loop.F, forward_step.F and pkg/exf/EXF_OPTIONS.h : |
83 |
|
- changes to use model/src/the_main_loop.F rather than the ecco version |
84 |
|
in ecco type adjoint runs. |
85 |
|
o model/src: |
86 |
|
- move computation of recip of grid distance & area from INI_MASKS_ETC to INI_GRID |
87 |
|
- move setting of h0Fac[C,W,S] from INI_LINEAR_PHISURF to INI_MASKS_ETC |
88 |
|
- update_masks_etc.F: rename SMOOTH*_R4,R8 function to the corresponding type (RS,RL) |
89 |
|
o model/src: |
90 |
|
- fix S/R UPDATE_CG2D for ALLOW_DEPTH_CONTROL case (was empty before); |
91 |
|
update output of test experiment bottom_ctrl_5x5. |
92 |
|
o pkg/generic_advdiff: |
93 |
|
- implement vertical bi-harmonic diffusion (for now, only for T & S and |
94 |
|
only explicitly). |
95 |
|
o verification/bottom_ctrl_5x5/ |
96 |
|
update reference output (results changed quite some time ago, |
97 |
|
not sure when; but new reference results look ok in terms of grdchk) |
98 |
|
o verification/1D_ocean_ice_column/ |
99 |
|
For verification purposes, the exact adjoint rather than an approx. |
100 |
|
should be used (e.g., to comare against F.D. or TLM). |
101 |
|
removed autodiff_inadmode*_ad.F routines in code_ad/ |
102 |
|
Update results/ accordingly. |
103 |
|
o pkg/cal: |
104 |
|
- change cal_getdate.F so that output date is function of input time only |
105 |
|
(less ambiguous, not mixed up with input iteration number). |
106 |
|
- fix calendarDumps for non-zero phase-lag (S/R CAL_TIME2DUMP). |
107 |
|
o pkg/diagnostics: |
108 |
|
- implement calendarDumps also for stats-diags |
109 |
|
- implement calendarDumps also for snap-shot output (negative freq) |
110 |
|
- for now, ignores calendarDumps if non-zero phase-lag. |
111 |
|
|
112 |
|
checkpoint63n (2012/06/04) |
113 |
|
o model/src: |
114 |
|
- recompute rSurfW & S to ensure consistent column thickness (needed for rStar) |
115 |
|
between Sum_k(hFac*drF) and rSurf-rLow, for now only when using pkg/shelfice; |
116 |
|
- use h0FacC, W & S to compute hFac_surfC, W & S in calc_surf_dr.F (more |
117 |
|
similar to calc_r_star.F); update results of experiment internal_wave and |
118 |
|
tutorial_global_oce_in_p (affected @ truncation level). |
119 |
|
o pkg/shelfice: |
120 |
|
- add parameterisation of advective-diffusive flux into the ice shelf |
121 |
|
according to Holland and Jenkins (1999), eq.22-33, as suggested by Jan De Ryd |
122 |
|
- simply eps3/3a computation |
123 |
o eesupp: |
o eesupp: |
124 |
- add new CPP option (DISCONNECTED_TILES) to disconnect tiles (no exchange |
- add new CPP option (DISCONNECTED_TILES) to disconnect tiles (no exchange |
125 |
between tiles, just fill-in edges assuming locally periodic subdomain) and |
between tiles, just fill-in edges assuming locally periodic subdomain) and |