3 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
4 |
============================== |
============================== |
5 |
|
|
6 |
|
o corrected CPP option mismatch in exf_Bulk... functions for case |
7 |
|
#define ALLOW_ATM_WIND && #undef ALLOW_ATM_TEMP |
8 |
|
|
9 |
|
checkpoint50a_post |
10 |
|
o Bug fix for merging between c50 and KPP. |
11 |
|
ikey was passed from thermodynamics to kpp_calc via |
12 |
|
common block rather than being recomputed in kpp_calc, |
13 |
|
in contradiction with new key itdkey. |
14 |
|
New key ikppkey created, and tamc.h headers updated. |
15 |
|
|
16 |
|
checkpoint50 |
17 |
|
- 2nd step of merge with ecco-branch: |
18 |
|
merge changes in ecco-branch back into main trunk |
19 |
|
- bug fix in find_alpha |
20 |
|
- more storing in gmredi_calc_tensor.F, kpp_routines.F |
21 |
|
- avoid iterative loop in ini_pressure for ALLOW_AUTODIFF_TAMC |
22 |
|
- merged some sliced AD support routines |
23 |
|
|
24 |
|
checkpoint49 |
25 |
|
merged from ecco-branch: |
26 |
|
o exf: |
27 |
|
- Enable initialisation of forcing fields to constant |
28 |
|
(runtime) values. |
29 |
|
- in exf_getffields.F |
30 |
|
Reduce i-/j-loop to interior domain, discarding overlaps. |
31 |
|
That also fixes wrong TAF-key computations for key_1, key_2 |
32 |
|
with bulf formulae. |
33 |
|
- exf_init.F modify #ifdef for exf_init_evap |
34 |
|
- exf_getffieldrec.F, ctrl_getrec.F |
35 |
|
The following INT-usages are not safe: |
36 |
|
fldsecs = int(fldsecs/fldperiod)*fldperiod |
37 |
|
fldcount = int(fldsecs/fldperiod) + 1 |
38 |
|
and were modified. |
39 |
|
o autodiff: checkpoint_lev... |
40 |
|
- updated storing for bulk formulae (some arrays were missing) |
41 |
|
- added state for seaice package |
42 |
|
- synchronized with main branch |
43 |
|
o DIVA: |
44 |
|
Modif.s in |
45 |
|
- the_main_loop.F |
46 |
|
- the_model_main.F |
47 |
|
and added cost_final_restore.F cost_final_store.F |
48 |
|
o With these changes, it is possible to fully separate |
49 |
|
the first forward run to compute the cost function (mdthe_main_loop) |
50 |
|
and adjoint runs over individual DIVA intervals (adthe_main_loop) |
51 |
|
into separate model executions. |
52 |
|
This enables better tuning of DIVA to given queue sizes. |
53 |
|
o Loss of cost function value in consecutive DIVA runs is |
54 |
|
fixed by storing to 'costfinal' file. |
55 |
|
o Appropriate call of ctrl_unpack/_pack,grdchk_main at |
56 |
|
beginning/end of adjoint integration is accounted for. |
57 |
|
The current changes apply to TAF-1.4.26. |
58 |
|
TAF has one remaining bug, requiring one manual change |
59 |
|
(see MITgcm manual). |
60 |
|
o cal: synchronized between MAIN and ecco-branch |
61 |
|
|
62 |
|
checkpoint48i_post |
63 |
|
o Changes to restore differentiability of code w.r.t. previous tag |
64 |
|
(mostly adding new routines to make list and replacing |
65 |
|
pressure by totPhiHyd). |
66 |
|
o Updated and merged grdchk package |
67 |
|
- has both ADM and TLM checks |
68 |
|
- works for single- and multi-proc. |
69 |
|
- output cleaned |
70 |
|
- worked successfully for parallel DIVA |
71 |
|
o genmake: |
72 |
|
- Adding -platform=SP4 |
73 |
|
- Modif. for DEC Alpha (many don't have makedepend; use mkdep instead). |
74 |
|
o Modif. for DIVA: |
75 |
|
ctrl_unpack should not be called in intermediate DIVA interval. |
76 |
|
(similar thing would apply for active files such as |
77 |
|
tbar, sbar, psbar, which should not be reinitialised; |
78 |
|
only relevant for ecco-branch) |
79 |
|
|
80 |
|
checkpoint48h_post |
81 |
|
o improve "ini_p_ground" when using finite difference form for calc_phi_hyd |
82 |
|
o Added net flux to downward flux conversion to pkg/exf/exf_getffields.F |
83 |
|
o Added SEAICE_initialHEFF to pkg/seaice |
84 |
|
o ecco_the_model_main.F: SEAICE_WRITE_PICKUP must precede WRITE_CHECKPOINT |
85 |
|
|
86 |
|
checkpoint48g_post |
87 |
|
--- EOS: |
88 |
|
o compute locally the pressure for use in EOS: UNESCO, JMD95P, MDJWF |
89 |
|
store total Potential in totPhyHyd for diagnostics & EOS. |
90 |
|
o fix restart and overlap Pb when using Z-coord and EOS funct. of P |
91 |
|
--- Pressure/geopotential gradient: |
92 |
|
o use Ro_surf & R_low instead of hFac in cal_phi_hyd: |
93 |
|
o NLFS + p-coord.: eta*Alpha' is not longer lagging 1.time-step behind |
94 |
|
( changes results of global_ocean_pressure ) |
95 |
|
o NLFS + z-coord.: add eta*Rho' contribution. |
96 |
|
o accurate phiHyd when using interface-W at the middle between 2 centers |
97 |
|
( changes results of ideal_2D_oce ) |
98 |
|
o includes r* 2nd term in potential gradient. |
99 |
|
|
100 |
|
checkpoint48f_post |
101 |
|
Merging from release1_p12: |
102 |
|
o Modifications for using pkg/exf with pkg/seaice |
103 |
|
- improved description of the various forcing configurations |
104 |
|
- added basic radiation bulk formulae to pkg/exf |
105 |
|
- units/sign fix for evap computation in exf_getffields.F |
106 |
|
- updated verification/global_with_exf/results/output.txt |
107 |
|
o Added pkg/sbo for computing IERS Special Bureau for the Oceans |
108 |
|
(SBO) core products, including oceanic mass, center-of-mass, |
109 |
|
angular, and bottom pressure (see pkg/sbo/README.sbo). |
110 |
|
o Lower bound for viscosity/diffusivity in pkg/kpp/kpp_routines.F |
111 |
|
to avoid negative values in shallow regions. |
112 |
|
- updated verification/natl_box/results/output.txt |
113 |
|
- updated verification/lab_sea/results/output.txt |
114 |
|
o MPI gather, scatter: eesupp/src/gather_2d.F and scatter_2d.F |
115 |
|
o Added useSingleCpuIO option (see PARAMS.h). |
116 |
|
o Updated useSingleCpuIO option in mdsio_writefield.F to |
117 |
|
work with multi-field files, e.g., for single-file pickup. |
118 |
|
o pkg/seaice: |
119 |
|
- bug fix in growth.F: QNET for no shortwave case |
120 |
|
- added HeffFile for specifying initial sea-ice thickness |
121 |
|
- changed SEAICE_EXTERNAL_FLUXES wind stress implementation |
122 |
|
o Added missing /* */ to CPP comments in pkg/seaice, pkg/exf, |
123 |
|
kpp_transport_t.F, forward_step.F, and the_main_loop.F |
124 |
|
o pkg/seaice: |
125 |
|
- adjoint-friendly modifications |
126 |
|
- added a SEAICE_WRITE_PICKUP at end of the_model_main.F |
127 |
|
|
128 |
|
checkpoint48e_post |
129 |
|
o New S/R for diagnostic of bottom pressure (phi0surf was missing in c48d_post) |
130 |
|
o dynamics: change definition of computational domain & adapt mom_fluxform |
131 |
|
accordingly ; when viscA4=0, allows to run the dynamics with Olx=Oly=2. |
132 |
|
o following exp works with Olx=Oly=2 (and pass through checkbounds option): |
133 |
|
adjust_nlfs.cs-32x32x1 adjustment.128x64x1 adjustment.cs-32x32x1 |
134 |
|
exp0 exp1 exp2 exp4 exp5 hs94.cs-32x32x5 hs94.128x64x5 hs94.1x64x5 |
135 |
|
global_ocean.90x40x15 global_with_exf global_ocean_pressure |
136 |
|
inverted_barometer solid-body.cs-32x32x1 |
137 |
|
(high order AdvScheme and/or multiDimAd works with Olx=Oly=2 ? not sure) |
138 |
|
|
139 |
|
checkpoint48d_post |
140 |
|
o in preparation for r*: |
141 |
|
a) use pre-computed gradient of hydrostatic potential: |
142 |
|
changes in timestep.F & mom_cdscheme.F affects results of ideal_2D_oce |
143 |
|
b) move phi0surf from calc_phi_hyd to calc_grad_phi_hyd: |
144 |
|
=> affects results of glob_oce_pressure (different truncation error) |
145 |
|
update output.txt of ideal_2D_oce & global_ocean_pressure |
146 |
|
|
147 |
|
checkpoint48d_pre |
148 |
|
o in preparation for r*: |
149 |
|
new S/R (calc_grad_phi_hyd.F) to compute Hydrostatic potential gradient. |
150 |
|
pass the 2 comp. of the grad. as arguments to momentum S/R. |
151 |
|
but for the moment, only used if it does not change the results. |
152 |
|
o add finite volume form of calc_phi_hyd for buoyancyRelation OCEANIC & OCEANICP |
153 |
|
|
154 |
|
checkpoint48c_post |
155 |
|
Modifications in treatment of initial flux forcing controls |
156 |
|
to avoid extra storing of flux fields. |
157 |
|
|
158 |
|
checkpoint48c_pre |
159 |
o Added tidal_basin_2d experiment (currently 1D!) |
o Added tidal_basin_2d experiment (currently 1D!) |
160 |
o can use r* Coordinate with OBC (but Orlanski Rad. BC not yet implemented) |
o can use r* Coordinate with OBC (but Orlanski Rad. BC not yet implemented) |
161 |
|
|