2 |
|
|
3 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
4 |
============================== |
============================== |
5 |
|
o use recip_dx*,recip_dy* instead of /dx*,/dy* in orlanski_E,W,N,S |
6 |
|
(affects plume_on_slope "cg2d" output) |
7 |
|
o NONLIN_FRSURF: rescale gNm1 to get a better conservation with A-B |
8 |
|
o change phiMin in exp1 (to agree with documentation) |
9 |
|
o add new flags: |
10 |
|
- T,S forcing inside/outside Adams-Bashforth |
11 |
|
- turn off temp,salt advection and/or forcing |
12 |
|
- for each tracer, define internal flag for multiDimAdvection & A-B |
13 |
|
o monitor: change definition of KE to check conservation of Energy |
14 |
|
|
15 |
|
checkpoint45c_post |
16 |
|
o Removed f90 comments ("!") |
17 |
|
o Modified initialisations to enable exf package for MITgcm |
18 |
|
(without invoking ECCO_PACKAGE). |
19 |
|
o modifications to gradient check package (Martin Losch) |
20 |
|
- enable centered differences |
21 |
|
- modified format of standard output |
22 |
|
|
23 |
|
checkpoint45b_post |
24 |
|
o cleaned exchange calls in cg2d |
25 |
|
o Included CPP option ALLOW_AUTODIFF_MONITOR in forward_step |
26 |
|
o included CPP option SINGLE_LAYER_MODE in thermodynamics |
27 |
|
to configure barotropic setup (Martin Losch) |
28 |
|
o moved some initialisations within bi,bj loop in dynamics |
29 |
|
(TAF flow dependency) |
30 |
|
o in initialise_varia, moved call packages_init_variables |
31 |
|
before convective_adjustment_init |
32 |
|
(corresponding adjustment of TAF tape initialisation). |
33 |
|
o in calc_diffusivity/calc_viscosity extended domain to full overlap. |
34 |
|
|
35 |
|
checkpoint45a_post |
36 |
|
o compute Rho over the whole domain (extend iMin,jMin indices) |
37 |
|
o specific exchange for X-slice domain (case Ny=1) |
38 |
|
o argument futureIter of S/R obcs_calc missing in exp4 & internal_wave /code |
39 |
|
o solve_for_pressure : set cg2d_x to zero where OBCS are applied |
40 |
|
|
41 |
|
checkpoint44h_post = checkpoint45 |
42 |
|
o Fix to restore differentiability: |
43 |
|
place at which packages_init_variables is called in |
44 |
|
initialise_varia is crucial for reverse control flow. |
45 |
|
o modifications in GMRedi to enable stable adjoint |
46 |
|
(corresponding modif's in thermodynamics, dynamics) |
47 |
|
o added missing hook for sponge layer code in external_forcing |
48 |
|
o modified test cost functions |
49 |
|
o storing in gad_advection (plus removal of write(0,*)! |
50 |
|
o missing headers TR1.h added in convective_adjustment routines |
51 |
|
o errorMessageUnit set non-equal zero in eeboot |
52 |
|
(conflict on some platforms, e.g. IBM SP3) |
53 |
|
o modified carbon verif. |
54 |
|
|
55 |
|
checkpoint44h_pre |
56 |
|
o vertical grid option: allow to put the Interface at the middle between |
57 |
|
2 cell-centers ; replace delZ in pkg/kpp by drF. |
58 |
|
o GM Advective form: Tracers are advected using the residual transport (= |
59 |
|
Euler+GM-bolus); set param GM_AdvSeparate=T to return to previous form. |
60 |
|
|
61 |
|
checkpoint44g_post |
62 |
|
o fix surface correction term in multiDimAdvection (affects plume_on_slope |
63 |
|
results) ; ifdef missing in thermodynamics.F |
64 |
|
|
65 |
|
checkpoint44f_pre,checkpoint44f_post |
66 |
|
o added PTRACERS package |
67 |
|
This allows an arbitrary number of passive tracers to be integrated |
68 |
|
forward simultaneously with the dynamicaly model. |
69 |
|
+ Implemented so far: |
70 |
|
- basic forward algorithm (time-stepping, advection, diffusion, convection) |
71 |
|
- I/O and checkpointing |
72 |
|
- GM/Redi *but* using the GM/Redi coefficient of Salt |
73 |
|
+ Not implemented so far: |
74 |
|
- KPP |
75 |
|
- OBCS |
76 |
|
+ No specific example supplied (yet) but global_ocean.90x40x15 has the |
77 |
|
necessary data.ptracer file. Simply use -enable=ptracers and uncomment |
78 |
|
line in data.pkg. |
79 |
|
+ This package is disabled by default since it increases storage. |
80 |
|
|
81 |
|
|
82 |
|
o testscript: added -cleanup option |
83 |
|
o Shap_filt: a) new shap_filt S/R to use no-slip BC with S2 filter |
84 |
|
b) enable to filter 2D fields. |
85 |
|
|
86 |
|
checkpoint44e_pre,post |
87 |
|
o re-wrote convective adjustment to use pre-calculated weights |
88 |
|
Affects: model/src/convective_adjustment.F, convective_adjustment_ini.F |
89 |
|
Adds: model/src/convective_weights.F, convectively_mixtracer.F |
90 |
|
Original calls to convect() still in place. |
91 |
|
|
92 |
|
chkpt44d_pre,post |
93 |
|
o added missing EXCLUDE_MONITOR flags |
94 |
|
o changed "e" to "_d" in gmredi_slope_limit, gmredi_slope_psi |
95 |
|
(incompatible typ in MIN/MAX expressions caused problems |
96 |
|
on IBM SP3) |
97 |
|
o in genmake added variable MAKEDEPEND |
98 |
|
plus resetting for case SunOS |
99 |
|
o added timer_stats.c routine for IBM SP3 |
100 |
|
o removed variables in dynamics |
101 |
|
!!! I forgot to create tag chkpt44d_pre !!! |
102 |
|
!!! But chkpt44d_pre = chkpt44c_post !!! |
103 |
|
|
104 |
|
o real fresh water flux implemented with non-linear free-surface. |
105 |
|
|
106 |
|
chkpt44c_pre,post |
107 |
|
o few fix (mask in shap_s2, EmPmR in external_field_load, |
108 |
|
USE_NATURAL_BCS in solve_for_P); |
109 |
|
o add arguments myIter & myTime to S/R obcs_calc & solve_for_P |
110 |
|
|
111 |
|
checkpoint44b_pre/_post |
112 |
|
o merge of relevant stuff from the ecco-branch: |
113 |
|
- genmake: removed $S64 overwrite for case SunOS |
114 |
|
- pkg/exf: update and corrections for field swapping and obcs |
115 |
|
- pkg/ecco: parameter lists for the_model_main, the_main_loop |
116 |
|
harmonized between ECCO and MITgcm |
117 |
|
- pkg/autodiff: added flow directives for obcs, mdsio_gl_slice |
118 |
|
updated checkpointing_lev... lists for obcs |
119 |
|
- model/src: minor changes in forward_step, plot_field |
120 |
|
added directive for divided adjoint in the_main_loop |
121 |
|
- pkg/mdsio: added mdsio_gl_slice |
122 |
|
|
123 |
|
o check parameters & config (chkpt44a_pre,post) |
124 |
|
o OBC and NonLin_FrSurf. |
125 |
|
|
126 |
|
checkpoint44 |
127 |
|
o fix bug in mom_vi_del2uv |
128 |
|
o select when filters are applied ; add options to zonal_filter (data.zonfilt) |
129 |
|
o gmredi: fix Pb in the adiabatic form ; add options (.e.g. Bolus advection) |
130 |
|
o update AIM experiments (NCEP input files) |
131 |
|
o improve and extend diagnostics (Monitor, TimeAve with NonLin-FrSurf) |
132 |
|
o added some stuff for AD |
133 |
|
|
134 |
|
checkpoint43 |
135 |
|
o added two new advection tests |
136 |
|
o added global 4x4 degree ocean test |
137 |
|
o added test of H&S94 with free surface on cube |
138 |
|
o some final mods for gradient check, adjoint,... |
139 |
|
|
140 |
|
checkpoint42 |
141 |
|
Fixed AD-related problems: |
142 |
|
o Store directives up-to-date with re-arranged Adams-Bashforth |
143 |
|
(mainly thermodynamics.F) |
144 |
|
o New store directives for multi-dim. advection schemes |
145 |
|
* new CPP flag ALLOW_MULTI_DIM_ADVECTION |
146 |
|
* new common block and key passkey |
147 |
|
(mainly gad_advection.F) |
148 |
|
o Modified store directives for split of dynamics/thermodynamics |
149 |
|
for the case ALLOW_KPP |
150 |
|
o Cleaned argument list for timestep_tracer.F |
151 |
|
|
152 |
|
checkpoint41 |
153 |
|
o re-formatted/added comments in prologues of key subroutines |
154 |
|
to work with protex |
155 |
|
|
156 |
|
checkpoint40 |
157 |
|
o moved calc_gs, calc_gt and calc_tr1 back to model/src |
158 |
|
o added FLT (floats) package from Arne along with demo. experiment |
159 |
|
o re-arranged Adams-Bashforth scheme for tracers to allow easier |
160 |
|
implementation of forward-in-time forcing terms |
161 |
|
o more mods for non-linear free-surface including new variable |
162 |
|
o modified multi-dim method to work on cube |
163 |
|
o bug fix in KPP (from DM) |
164 |
|
|
165 |
|
checkpoint40pre9 |
166 |
|
o added DST advection schemes |
167 |
|
o fix MPI bug on SGI's |
168 |
|
o modifications for non-linear free-surface |
169 |
|
o fixed conditionals for fixed flow/passive tracer only mode |
170 |
|
o added tension/strain form of stress tensor |
171 |
|
o changes to commenting of arguments for TAF/TAMC |
172 |
|
o added/updated many matlab scripts |
173 |
|
o added multi-dimensional advection method for tracers |
174 |
|
|
175 |
|
checkpoint40pre8 |
176 |
|
o new package rw/ |
177 |
|
Added method for dumping intermediate local arrays: |
178 |
|
mdsio_writetile - same as mdsio_writefield except works from inside |
179 |
|
bi,bj loop |
180 |
|
mdsio_writelocal - same as mdsio_writetile except works for local |
181 |
|
arrays |
182 |
|
write_local_r? - higher-level wrapper for mdsio_writelocal |
183 |
|
Controlled by diagFreq. Defaults to zero (ie. no dumps) |
184 |
|
Example given at end of mom_vecinv.F that dumps some local arrays. |
185 |
|
o replaced calc_mom_rhs.F by mom_fluxform.F/mom_vecinc.F to |
186 |
|
control flux-form versus vector invariant code. |
187 |
|
Switch is runtime flag vectorInvariantMomentum (defaults to false) |
188 |
|
or genmake -DISABLE ... |
189 |
|
o TAMC-related update to cope with mom_fluxform.F/mom_vecinc.F |
190 |
|
(additional storing is now needed for case vecinv; |
191 |
|
currently nothing implemented to switch this storing off |
192 |
|
in case fluxform is used since no CPP flag available). |
193 |
|
|
194 |
|
checkpoint40pre7 |
195 |
|
o bug fix in dynamics.F |
196 |
|
calc_viscosity called before k=1,Nr loop. |
197 |
|
o added some #ifdef's ALLOW_PASSIVE_TRACER |
198 |
|
o cleaned initialisations in ctrl_map_ini |
199 |
|
related to TAF |
200 |
|
|
201 |
|
checkpoint40pre6 |
202 |
|
o split dynamics into thermodynamics & dynamics |
203 |
|
o corresponding split of calc_diffusivity into |
204 |
|
calc_diffusivity, calc_viscosity |
205 |
|
(kpp_calc_diff, kpp_calc_visc). |
206 |
|
o Added diffkr, kapgm to control variables |
207 |
|
o bug fix in gmredi_slope_limit for ldd97 |
208 |
|
|
209 |
|
checkpoint40pre5 |
210 |
|
o proper initialization of Adams-Bashforth time stepping |
211 |
|
+ uses forward step for very first step of model (iter=0) |
212 |
|
o re-generated *all* output since almost all output is modified by this |
213 |
|
change (sorry but we've been putting this one off for about 2 years) |
214 |
|
+ natl_box, aim.5l_Equatorial_Channel and aim.5l_zon-ave were |
215 |
|
*not* regenerated since they are already out of date and I don't |
216 |
|
want to create the impression that they are suddenly working. |
217 |
|
o removed KAP stuff for DEC's |
218 |
|
o modified/added many .cvsignore files in */code and */input |
219 |
|
o added new expt "solid-body.cs-32x32x1" |
220 |
|
+ this uses an alternative form of ini_curvilinear_grid.F which |
221 |
|
is not necessarily any better than the current one... |
222 |
|
|
223 |
|
checkpoint40pre4 |
224 |
|
o added re-initialisations of utrans,vtrans,rtrans in |
225 |
|
calc_common_factors which are crucial for TAF. |
226 |
|
o extended iMin,jMin range for calc_common_factors, |
227 |
|
calc_diffusivity |
228 |
|
o added option ALLOW_PASSIVE_TRACER |
229 |
|
o a few minor updates and fixes |
230 |
|
|
231 |
|
checkpoint40pre3 |
232 |
|
o created calc_gtr1.F for gad package |
233 |
|
o split the_main_loop and re-introduced forward_step |
234 |
|
o some recomputations of nIter0 for TAMC |
235 |
|
|
236 |
|
checkpoint40pre2 |
237 |
|
o Updated to latest branch of AJA, JMC |
238 |
|
o Added gradient check package |
239 |
|
o Added routines for passive tracer handling |
240 |
|
o Added carbon sequestration sensitivity experiment |
241 |
|
|
242 |
|
checkpoint39 |
243 |
|
o Modifications and fixes to support TAMC |
244 |
|
(mainly missing or wrong directives) |
245 |
|
o Corrected common blocks when fooling TAMC to produce adjoint I/O |
246 |
|
in addummy_in_stepping |
247 |
|
o Updated verification/vero/ (bug fixes in code/, correct pickups in input/) |
248 |
|
o Enabled "sliced" (xz-,yz-) exchanges, needed for obcs. |
249 |
|
o Included following packages: |
250 |
|
cal: calendar |
251 |
|
exf: external forcing |
252 |
|
ecco: essential routines to control use of cal, exf |
253 |
|
(NB: this is not the full ECCO environment) |
254 |
|
o disabled some packages by default in genmake |
255 |
|
|
256 |
|
checkpoint38 |
257 |
|
o see notes_c37_adj.txt |
258 |
|
o corrected wrong layer index in external_forcing_t |
259 |
|
o replaced the_main_loop/forward_step by the_main_loop |
260 |
|
o removed forceoutput from write_state |
261 |
|
(replaced criteria for initial/final state) |
262 |
|
o added runtime flag for pre-C35 pickups: usePickupBeforeC35 |
263 |
|
|
264 |
|
checkpoint37 |
265 |
|
o mdsio.F routine split and packaged in pkg/mdsio |
266 |
|
(mdsio.F and mdsio_gl.F still exist in eesupp/src but are not |
267 |
|
used unless genmake -disable=mdsio) |
268 |
|
o allow a more precise surface-geopotential gradient for the atmosphere |
269 |
|
taking into account the change in space of Alpha (=1/rho) at the surface. |
270 |
|
o variable "etaN" (replace cg2d_x) is the surface R-anomaly [r unit] |
271 |
|
(ocean: sea surface height [m] / atmos: surface pressure anomaly [Pa]) |
272 |
|
o Change units: Phi_Hydrostatic (=phiHyd), Phi_Surface(=cg2d_x) and |
273 |
|
Phi_NonHydrostatic (=cg3d_x), all have now the dimension of a potential |
274 |
|
[m2/s2] (pressure/rhoConst in the ocean ; Geo-Potential in the atmosphere). |
275 |
|
o Time-Average diagnostics defined as a pseudo package "timeave". |
276 |
|
o pickup files : old pickup (before c35) did not work with new code (c35 |
277 |
|
and after) unless activating 2 commented lines in initialise-varia. |
278 |
|
o Non-Hydrostatic pickup file Pb fixed. |
279 |
|
o bug with Rigid-lid version (since c35) fixed. |
280 |
|
|
281 |
|
checkpoint36 |
282 |
|
o implement Crank-Nickelson barotropic time-stepping |
283 |
|
o recover Time-Average diagnostics |
284 |
|
o fix Pb in checkpoint (PR#19) |
285 |
|
o AIM become a standard package ; clean unused CPP_kees |
286 |
|
|
287 |
|
checkpoint35 |
288 |
|
o subtantial rewrite of dynamics.f |
289 |
|
o allows staggered time-stepping |
290 |
|
o packaged obcs, aim (Atmospheric Physics), zonal_filt, shap_filt |
291 |
|
o added bottom drag (linear + quadratic) |
292 |
|
|
293 |
checkpoint34 |
checkpoint34 |
294 |
o Modified exchange routines for correct reverse simulation. |
o Modified exchange routines for correct reverse simulation. |