3 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
4 |
============================== |
============================== |
5 |
|
|
6 |
|
checkpoint45c_post |
7 |
|
o Removed f90 comments ("!") |
8 |
|
o Modified initialisations to enable exf package for MITgcm |
9 |
|
(without invoking ECCO_PACKAGE). |
10 |
|
o modifications to gradient check package (Martin Losch) |
11 |
|
- enable centered differences |
12 |
|
- modified format of standard output |
13 |
|
|
14 |
|
checkpoint45b_post |
15 |
|
o cleaned exchange calls in cg2d |
16 |
|
o Included CPP option ALLOW_AUTODIFF_MONITOR in forward_step |
17 |
|
o included CPP option SINGLE_LAYER_MODE in thermodynamics |
18 |
|
to configure barotropic setup (Martin Losch) |
19 |
|
o moved some initialisations within bi,bj loop in dynamics |
20 |
|
(TAF flow dependency) |
21 |
|
o in initialise_varia, moved call packages_init_variables |
22 |
|
before convective_adjustment_init |
23 |
|
(corresponding adjustment of TAF tape initialisation). |
24 |
|
o in calc_diffusivity/calc_viscosity extended domain to full overlap. |
25 |
|
|
26 |
|
checkpoint45a_post |
27 |
|
o compute Rho over the whole domain (extend iMin,jMin indices) |
28 |
|
o specific exchange for X-slice domain (case Ny=1) |
29 |
|
o argument futureIter of S/R obcs_calc missing in exp4 & internal_wave /code |
30 |
|
o solve_for_pressure : set cg2d_x to zero where OBCS are applied |
31 |
|
|
32 |
|
checkpoint44h_post = checkpoint45 |
33 |
|
o Fix to restore differentiability: |
34 |
|
place at which packages_init_variables is called in |
35 |
|
initialise_varia is crucial for reverse control flow. |
36 |
|
o modifications in GMRedi to enable stable adjoint |
37 |
|
(corresponding modif's in thermodynamics, dynamics) |
38 |
|
o added missing hook for sponge layer code in external_forcing |
39 |
|
o modified test cost functions |
40 |
|
o storing in gad_advection (plus removal of write(0,*)! |
41 |
|
o missing headers TR1.h added in convective_adjustment routines |
42 |
|
o errorMessageUnit set non-equal zero in eeboot |
43 |
|
(conflict on some platforms, e.g. IBM SP3) |
44 |
|
o modified carbon verif. |
45 |
|
|
46 |
|
checkpoint44h_pre |
47 |
|
o vertical grid option: allow to put the Interface at the middle between |
48 |
|
2 cell-centers ; replace delZ in pkg/kpp by drF. |
49 |
|
o GM Advective form: Tracers are advected using the residual transport (= |
50 |
|
Euler+GM-bolus); set param GM_AdvSeparate=T to return to previous form. |
51 |
|
|
52 |
|
checkpoint44g_post |
53 |
|
o fix surface correction term in multiDimAdvection (affects plume_on_slope |
54 |
|
results) ; ifdef missing in thermodynamics.F |
55 |
|
|
56 |
|
checkpoint44f_pre,checkpoint44f_post |
57 |
|
o added PTRACERS package |
58 |
|
This allows an arbitrary number of passive tracers to be integrated |
59 |
|
forward simultaneously with the dynamicaly model. |
60 |
|
+ Implemented so far: |
61 |
|
- basic forward algorithm (time-stepping, advection, diffusion, convection) |
62 |
|
- I/O and checkpointing |
63 |
|
- GM/Redi *but* using the GM/Redi coefficient of Salt |
64 |
|
+ Not implemented so far: |
65 |
|
- KPP |
66 |
|
- OBCS |
67 |
|
+ No specific example supplied (yet) but global_ocean.90x40x15 has the |
68 |
|
necessary data.ptracer file. Simply use -enable=ptracers and uncomment |
69 |
|
line in data.pkg. |
70 |
|
+ This package is disabled by default since it increases storage. |
71 |
|
|
72 |
|
|
73 |
|
o testscript: added -cleanup option |
74 |
|
o Shap_filt: a) new shap_filt S/R to use no-slip BC with S2 filter |
75 |
|
b) enable to filter 2D fields. |
76 |
|
|
77 |
|
checkpoint44e_pre,post |
78 |
|
o re-wrote convective adjustment to use pre-calculated weights |
79 |
|
Affects: model/src/convective_adjustment.F, convective_adjustment_ini.F |
80 |
|
Adds: model/src/convective_weights.F, convectively_mixtracer.F |
81 |
|
Original calls to convect() still in place. |
82 |
|
|
83 |
|
chkpt44d_pre,post |
84 |
|
o added missing EXCLUDE_MONITOR flags |
85 |
|
o changed "e" to "_d" in gmredi_slope_limit, gmredi_slope_psi |
86 |
|
(incompatible typ in MIN/MAX expressions caused problems |
87 |
|
on IBM SP3) |
88 |
|
o in genmake added variable MAKEDEPEND |
89 |
|
plus resetting for case SunOS |
90 |
|
o added timer_stats.c routine for IBM SP3 |
91 |
|
o removed variables in dynamics |
92 |
|
!!! I forgot to create tag chkpt44d_pre !!! |
93 |
|
!!! But chkpt44d_pre = chkpt44c_post !!! |
94 |
|
|
95 |
|
o real fresh water flux implemented with non-linear free-surface. |
96 |
|
|
97 |
|
chkpt44c_pre,post |
98 |
|
o few fix (mask in shap_s2, EmPmR in external_field_load, |
99 |
|
USE_NATURAL_BCS in solve_for_P); |
100 |
|
o add arguments myIter & myTime to S/R obcs_calc & solve_for_P |
101 |
|
|
102 |
|
checkpoint44b_pre/_post |
103 |
|
o merge of relevant stuff from the ecco-branch: |
104 |
|
- genmake: removed $S64 overwrite for case SunOS |
105 |
|
- pkg/exf: update and corrections for field swapping and obcs |
106 |
|
- pkg/ecco: parameter lists for the_model_main, the_main_loop |
107 |
|
harmonized between ECCO and MITgcm |
108 |
|
- pkg/autodiff: added flow directives for obcs, mdsio_gl_slice |
109 |
|
updated checkpointing_lev... lists for obcs |
110 |
|
- model/src: minor changes in forward_step, plot_field |
111 |
|
added directive for divided adjoint in the_main_loop |
112 |
|
- pkg/mdsio: added mdsio_gl_slice |
113 |
|
|
114 |
|
o check parameters & config (chkpt44a_pre,post) |
115 |
|
o OBC and NonLin_FrSurf. |
116 |
|
|
117 |
|
checkpoint44 |
118 |
|
o fix bug in mom_vi_del2uv |
119 |
|
o select when filters are applied ; add options to zonal_filter (data.zonfilt) |
120 |
|
o gmredi: fix Pb in the adiabatic form ; add options (.e.g. Bolus advection) |
121 |
|
o update AIM experiments (NCEP input files) |
122 |
|
o improve and extend diagnostics (Monitor, TimeAve with NonLin-FrSurf) |
123 |
|
o added some stuff for AD |
124 |
|
|
125 |
|
checkpoint43 |
126 |
|
o added two new advection tests |
127 |
|
o added global 4x4 degree ocean test |
128 |
|
o added test of H&S94 with free surface on cube |
129 |
|
o some final mods for gradient check, adjoint,... |
130 |
|
|
131 |
|
checkpoint42 |
132 |
|
Fixed AD-related problems: |
133 |
|
o Store directives up-to-date with re-arranged Adams-Bashforth |
134 |
|
(mainly thermodynamics.F) |
135 |
|
o New store directives for multi-dim. advection schemes |
136 |
|
* new CPP flag ALLOW_MULTI_DIM_ADVECTION |
137 |
|
* new common block and key passkey |
138 |
|
(mainly gad_advection.F) |
139 |
|
o Modified store directives for split of dynamics/thermodynamics |
140 |
|
for the case ALLOW_KPP |
141 |
|
o Cleaned argument list for timestep_tracer.F |
142 |
|
|
143 |
|
checkpoint41 |
144 |
|
o re-formatted/added comments in prologues of key subroutines |
145 |
|
to work with protex |
146 |
|
|
147 |
checkpoint40 |
checkpoint40 |
148 |
o moved calc_gs, calc_gt and calc_tr1 back to model/src |
o moved calc_gs, calc_gt and calc_tr1 back to model/src |
149 |
o added FLT (floats) package from Arne along with demo. experiment |
o added FLT (floats) package from Arne along with demo. experiment |