2 |
|
|
3 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
4 |
============================== |
============================== |
5 |
|
o aim.5l_cs: modify sea-ice fraction input file ; update output |
6 |
|
|
7 |
|
checkpoint52k_post |
8 |
|
o removed top-level directory "diags" |
9 |
|
- we migrated these functions to pkg/timeave a long time ago |
10 |
|
o fixed bug introduced in pkg/monitor |
11 |
|
o accuracy ctrlprec = 32 insuffient for gradient checks using |
12 |
|
averaged fields (I/O via cost_averages) |
13 |
|
-> use ctrl.h in active_file*.F to control I/O precision. |
14 |
|
o paramter list of CALL ADTHE_MAIN_LOOP |
15 |
|
o added initial do_the_model_io to ecco |
16 |
|
o set all ...Freq to zero after cost_final |
17 |
|
taveFreq, dumpFreq, pChkptFreq |
18 |
|
o MAIN and ECCO differentiability restored |
19 |
|
|
20 |
|
checkpoint52j_post |
21 |
|
o fix nchklev in tamc.h for global_ocean... |
22 |
|
o Bringing flow and store directives up-to-date |
23 |
|
o Really weird thing! |
24 |
|
call do_the_model_io from the_main_loop with parameters |
25 |
|
starttime, niter0, instead of mytime, myiter, screws up the adjoint: |
26 |
|
TAF places a faulty re-call of initialise_varia |
27 |
|
after cost_final call which should not be there! |
28 |
|
checkpoint52j_pre |
29 |
|
|
30 |
|
o tweaked pkg/mdsio/mdsio_readfield.F to have more useful logic and |
31 |
|
error messages |
32 |
|
o fix errorMessageUnit problem. remove old version of mdsio from eesup/src |
33 |
|
o modified pkg/mdsio/mdsio_readfield.F -- commented out diskspace-eating |
34 |
|
debug output |
35 |
|
o modfied mdsio_writefield.F to handle non-global files under exch2 |
36 |
|
o update AIM Cubed-Sphere exp (aim.5l_cs): |
37 |
|
use land, p*, and Franco.M forcing fields with fractional land-sea mask. |
38 |
|
o update AIM Equatorial Channel exp: |
39 |
|
use standard aim_v23 pkg and change the SST fields to be more realistic. |
40 |
|
o globalFile now works with exch2 |
41 |
|
o merged mom_vi_calc_relvort3.F, mom_vi_del2uv.F and |
42 |
|
mon_vort3.F to make exch2 friendly with >1 tile per cube face |
43 |
|
o limit timeave output for hi-res integrations |
44 |
|
|
45 |
|
checkpoint52i_post |
46 |
|
o removed CPP_EEOPTIONS.h from verification/*/code |
47 |
|
o updated the tiling for global_ocean_pressure |
48 |
|
checkpoint52i_pre |
49 |
|
|
50 |
|
o update aim.5l_LatLon using the standard aim_v23 pkg (instead of |
51 |
|
the old pkg/aim). |
52 |
|
|
53 |
|
checkpoint52h_pre |
54 |
|
o shortened offending lines in exch2_send_rx?.template |
55 |
|
o modified W2_EXCH2_TOPOLOGY.h to reflect current topology generator |
56 |
|
o genmake2: add build & execute test for NetCDF code |
57 |
|
o modded aim.5l_cs and global_ocean.cs32x15 to use exch2, removed |
58 |
|
old grid files, moved CPP_EEMACROS.h to pkg/exch2 |
59 |
|
o added useSingleCPUIO flag to monitor.F and mon_out.F |
60 |
|
o added temporary hack for sea-ice dynamics on the cube: |
61 |
|
metric terms are ignored and it is assumed that |
62 |
|
there is no sea-ice on cube faces 1, 2, 4, and 5. |
63 |
|
|
64 |
|
checkpoint52f_post |
65 |
|
o added exch2 cube routines to pkg/exch2 |
66 |
|
o modified hs94.cs-32x32x5 to use it |
67 |
|
o added new grid files to hs94.cs-32x32x5 |
68 |
|
checkpoint52f_pre |
69 |
|
|
70 |
|
checkpoint52e_post |
71 |
|
o first attempt to solve vertical advection (momemtum & T,S) Implicitly |
72 |
|
for now, only implemented with Lax-Wendroff (tracer) and VectInv (mom) |
73 |
|
o step.1 : rewrite (as in MultiDimAdvec) explicit tracer stepping (gad_calc_rhs.F) |
74 |
|
to work with implicit vertical advection and AB ; |
75 |
|
o Implic. vert. advect. implemented with flux-limiter, c2, u3 and c4 scheme. |
76 |
|
o update output (gad_calc_rhs.F changes affect truncation error) of exp: |
77 |
|
front_relax, global_ocean .cs32x15 & _pressure, internal_wave, lab_sea, natl_box |
78 |
|
checkpoint52e_pre |
79 |
|
|
80 |
|
o fix bug in impldiff when Nr=1 or Nr=2. |
81 |
|
o change the albedo in pkg/thsice (as GISS model); update output_thsice.txt |
82 |
|
o little changes (fftpack, aim/phy_suflux ...) to pass the "checkbounds" test. |
83 |
|
o fix deltaTfreesurf in update_cg2d.F (left from c45d_post): affects 2 |
84 |
|
ocean exp. that use NLFS: global_ocean.cs32x15 & global_ocean_pressure |
85 |
|
o added exf_offset_atemp for Celsius to Kelvin conversion of atemp |
86 |
|
o eesupp/inc/EEPARAMS.h: MAX_NO_PROCS=1024 |
87 |
|
o pkg/cal/cal_toseconds.F fix for long (>60 years) integrations |
88 |
|
|
89 |
|
checkpoint52d_post |
90 |
|
o re-tiled many expts to allow comparison in multi-processor mode |
91 |
|
- updated output accordingly |
92 |
|
- changed hs94.128x64x5 to use intial T from file rather than use the |
93 |
|
random number generator |
94 |
|
o add infrastructure in model routines for fizhi and gridalt packages |
95 |
|
o change global_ocean.cs32x15 because of Pb with GM on CS-grid. |
96 |
|
o changes to permit hi-res, cubed-sphere, configuration |
97 |
|
- added useSingleCpuIO capability to mdsio_readfield.F and exf_interp.F |
98 |
|
- added "#undef ALLOW_USE_MPI" support to eesupp/src/scatter_2d.F |
99 |
|
- added pkg/exf/exf_set_uv.F for on-the-fly interpolation and rotation |
100 |
|
of surface winds for the cube |
101 |
|
(A verification experiment, based on global_ocean.cs32x15, but with |
102 |
|
pkg/seaice turned on and with on-the-fly interpolation from the NCEP |
103 |
|
Gaussian grid is described in MITgcm_contrib/high_res_cube/README_ice, |
104 |
|
complete with example surface forcing files and matlab scripts to look |
105 |
|
at the output and compare it with that of global_ocean.cs32x15.) |
106 |
|
o packages pickup: $PKG_write_pickup called from packages_write_pickup ; |
107 |
|
but call to $PKG_read_pickup remains in $PKG_ini_vars: change CD_code |
108 |
|
to conform to the common rule. |
109 |
|
o check-in the first version of the coupler + interfaces for multi-components |
110 |
|
set-up (Chris's version, adapted to work on cubed-sphere grid); use MPI. |
111 |
|
- imply modification in eesupp (MPI_COMM_MODEL < MPI_COMM_WORLD). |
112 |
|
- atmosphere: use pkg aim_v23 ; no sea-ice for now. |
113 |
|
- ocean and atmos. share the same horizontal grid (no interpolation). |
114 |
|
- exchanges between component and coupler is done in forward_step |
115 |
|
o new verification experiment, cpl_aim+ocn: |
116 |
|
coupled ocean - AIM atmosphere on cs32 grid. |
117 |
|
|
118 |
|
checkpoint52d_pre |
119 |
|
o surface flux time-av diagnostic modified for NLFS & RealFreshWater |
120 |
|
o add call to AIM pendant S/R in external_forcing & external_fields_load.F |
121 |
|
|
122 |
|
checkpoint52c_post |
123 |
|
o Fixing OPTIONS files for use of exf: |
124 |
|
- ALLOW_ECCO not needed |
125 |
|
- ECCO_CPPOPTIONS.h should not be included from |
126 |
|
EXF_OPTIONS.h/CAL_OPTIONS.h, but from CPP_OPTIONS.h |
127 |
|
as has been the case in the past |
128 |
|
- All cal/exf related options are currently set in |
129 |
|
single header file. Should remain so until we |
130 |
|
agree on a consistent new policy. |
131 |
|
o Removed all CPP options for cal package. |
132 |
|
cal has no more CPP options. |
133 |
|
o modified STOREs in GAD_ADVECTION |
134 |
|
o corrected key comp. for passkey |
135 |
|
|
136 |
|
checkpoint52b_post |
137 |
|
o new version: "thsice" of thermodynamic sea-ice pkg: |
138 |
|
- only 1 call from "forward_step"; conserve energy, fresh-water & salt. |
139 |
|
- modify fluxes but do not change theta directly; |
140 |
|
- light cleaning of bulk_force pkg (still needed with thsice). |
141 |
|
- add forcing fields + data files in global_ocean.cs32x15/inp_thsice |
142 |
|
+ output_thsice.txt to test thsice & bulk_force pkgs. |
143 |
|
o put back grid & coriolis monitor output. |
144 |
|
checkpoint52b_pre |
145 |
|
|
146 |
|
o call "do_the_model_io" at the beginning (=> dump initial state) |
147 |
|
o use rhoConstFresh to define convertEmP2rUnit (was done only in ocean-P) |
148 |
|
o sort out interplay between tamc.h and PTRACERS_OPTIONS.h |
149 |
|
(need info on NUMBER_OF_PTRACERS) |
150 |
|
o missing PTRACERS headers in some routines |
151 |
|
o default tamc.h in pkg/autodiff supposed to crash |
152 |
|
to ensure proper customization |
153 |
|
|
154 |
|
checkpoint52a_post |
155 |
|
o modifications to make FREEZE (_SURFACE) flux visible to pkg/kpp |
156 |
|
- moved surfaceTendencyTice from pkg/seaice to main code |
157 |
|
- subroutine FREEZE_SURFACE now limits only surface temperature |
158 |
|
and is called in FORWARD_STEP (if allowFreezing=.TRUE.) |
159 |
|
=> new output.txt for global_ocean.cs32x15 & global_with_exf |
160 |
|
- keep the old version of FREEZE for backward compatibility; |
161 |
|
global_ocean.90x40x15 left unchanged using useOldFreezing=.TRUE. |
162 |
|
- corresponding modifs for TAF directives |
163 |
|
o added surface flux output variables to TIMEAVE_STATVARS |
164 |
|
o time-averaged output for pkg/ptracers |
165 |
|
o fix problem with ECCO-related divided adjoint |
166 |
|
snapshot file was missing some ad-variables, |
167 |
|
but no warning issued by TAF |
168 |
|
|
169 |
|
checkpoint52a_pre |
170 |
|
o some bug fixes for #undef REAL4_IS_SLOW |
171 |
|
|
172 |
|
checkpoint52 = ecco_c52_e35 |
173 |
|
o after testing adjoints of checkpoint51u_post |
174 |
|
|
175 |
|
checkpoint51u_post |
176 |
|
o this is the post-merge tag of the ecco-branch |
177 |
|
(needs another clean checkout and re-testing) |
178 |
|
o various natl_box_adjoint configs and results |
179 |
|
but no input fields |
180 |
|
o obcs adjoint not yet merged! |
181 |
|
o adding ECCO global 1x1 and 2x2 degree configs |
182 |
|
o updated/extended some store directives for seaice |
183 |
|
o pkg/ecco now containes ecco-specific part of cost function |
184 |
|
o top level routines the_main_loop, forward_step |
185 |
|
supersede those in model/src/ |
186 |
|
previous input data.cost now in data.ecco |
187 |
|
(new namelist ecco_cost_nml) |
188 |
|
o in ctrl/ cleaned some cross-dependencies and updated CPP options |
189 |
|
|
190 |
|
checkpoint51t_post |
191 |
|
o this is the pre-merge tag of the ecco-branch |
192 |
|
|
193 |
|
checkpoint51s_post |
194 |
|
o cleanup: removed various obsolete #define-s |
195 |
|
o pkg/grdchk |
196 |
|
- renamed 'ph-grd 3' to 'grad-res' |
197 |
|
- merged from ecco-branch |
198 |
|
(remaining bug fixes for obcs gradient checks) |
199 |
|
- additional high-precision output for testreport |
200 |
|
(grep for 'precision_grdchk_result') |
201 |
|
|
202 |
|
checkpoint51r_post |
203 |
|
o cleaned up or removed CPP_OPTIONS.h files |
204 |
|
o updated results for some of the tests to reflect the new |
205 |
|
optfiles (-ieee: -O0 instead of -O3) |
206 |
|
|
207 |
|
checkpoint51q_post |
208 |
|
o cd_code: |
209 |
|
- moved cd_scheme.F -> cd_code_scheme.F |
210 |
|
- separate read_checkpoint from cd_code_ini_vars.F |
211 |
|
- separated cd_code part from write_checkpoint |
212 |
|
- updated AD_SOURCE, generated .flow |
213 |
|
- added CD_CODE_VARS.h to the_main_loop |
214 |
|
- Moving cd_code specific store dir.s to pkg/cd_code |
215 |
|
o ctrl package totally restructured |
216 |
|
- pack/unpack now optional and decoupled from |
217 |
|
xx_/adxx_ I/O |
218 |
|
- ctrl_pack/unpack cleaned |
219 |
|
(new routines ctrl_init_ctrlvar.F, pkg/ctrl/ctrl_init_wet.F) |
220 |
|
- confined inclusion of AD_CONFIG.h to where necessary. |
221 |
|
o genmake2: changed AD dependencies from SRCFILES to AD_FILES |
222 |
|
|
223 |
|
checkpoint51p_post |
224 |
|
o create initial cd_code package |
225 |
|
- WARNING: the initialization of fields within packages |
226 |
|
is broken because INI_FIELDS (where READ_CHECKPOINT is |
227 |
|
called) is called *before* PACKAGES_INIT_VARIABLES. |
228 |
|
This seems to be backwards since |
229 |
|
1) its important to initially zero the entire field including |
230 |
|
the "halo" region |
231 |
|
2) zeroing needs to be done *before* reading pickup info |
232 |
|
and the pickup call might or might not happen in any |
233 |
|
particular run |
234 |
|
- this needs an overhaul for R2 |
235 |
|
o cleanup verification/*/code/GMREDI_OPTIONS.h |
236 |
|
o cleanup testreport |
237 |
|
|
238 |
|
checkpoint51o_post |
239 |
|
o bringing AD-related verif's up to date with genmake2 |
240 |
|
- carbon/ |
241 |
|
- global_ocean.90x40x15/ |
242 |
|
- hs.94.1x64x5/ |
243 |
|
- (front_relax/ ) |
244 |
|
o (more) consistent directory structure for AD-related verif.'s |
245 |
|
o cleaning ALLOW_GRADIENT_CHECK -> ALLOW_GRDCHK |
246 |
|
o cleaning some ALLOW_TANGENTLINEAR_RUN -> ALLOW_AUTODIFF |
247 |
|
o bug fix in find_alpha.F for MDJWF: |
248 |
|
- modif. to alpha = 1/D*( dN/dT - rho*dD/Dt) to account for |
249 |
|
change rho -> rho-rhoConst |
250 |
|
- replace call find_rho to find_rhonum |
251 |
|
o initialisation of rFlx extended to full array (required by TAF) |
252 |
|
and shifted to thermodynamics |
253 |
|
o removed PTRACERS.h in ptracers routine |
254 |
|
o added surfacetendencyPtr to S/R parameter list pracers_forcing |
255 |
|
o Correcting initialisations for fVerU, fVerV. |
256 |
|
o helping TAF to recognize dependency on kUp for |
257 |
|
fVerU, fVerV |
258 |
|
o Replacing delZ by delR in pkg/ctrl/ |
259 |
|
checkpoint51o_pre |
260 |
|
|
261 |
|
o undid all of cp51 pending some cleanups and discussion |
262 |
|
o added the [#include "AD_CONFIG.h"] statement to all files that need |
263 |
|
it for adjoint/tl #defines |
264 |
|
o re-worked the build logic in genmake2 to support AD_CONFIG.h |
265 |
|
o removed tools/genmake since it no longer works |
266 |
|
|
267 |
|
checkpoint51n_post |
268 |
|
o modifications to make FREEZE flux visible to pkg/kpp |
269 |
|
- moved surfaceTendencyTice from pkg/seaice to main code |
270 |
|
- FREEZE & EXTERNAL_FORCING_SURF moved to FORWARD_STEP |
271 |
|
- subroutine FREEZE now limits only surface temperature |
272 |
|
(this means new output.txt for global_ocean.90x40x15, |
273 |
|
global_ocean.cs32x15, and global_with_exf) |
274 |
|
o added surface flux output variables to TIMEAVE_STATVARS |
275 |
|
|
276 |
|
checkpoint51n_pre |
277 |
|
o added "#undef ALLOW_USE_MPI" support to eesupp/src/gather_2d.F |
278 |
|
o added [#include "AD_CONFIG.h"] to all files that use the following |
279 |
|
defines: |
280 |
|
- ALLOW_ADJOINT_RUN |
281 |
|
- ALLOW_TANGENTLINEAR_RUN |
282 |
|
- ALLOW_ECCO_OPTIMIZATION |
283 |
|
|
284 |
|
checkpoint51m_post |
285 |
|
o incorporating changes to get TAMC/TAF working in genmake2 |
286 |
|
- added new TAMC/TAF targets to the Makefile |
287 |
|
- various cleanups related to package handling |
288 |
|
- these changes should have no impact on the forward model |
289 |
|
|
290 |
|
checkpoint51l_post |
291 |
|
o changes to pkg/cal and pkg/exf that allow and speed-up long integrations |
292 |
|
- faster algorithm for cal_addtime.F |
293 |
|
- higher precision arithmetic, where needed |
294 |
|
- pkg/exf *startdate variables changed from calendar date to |
295 |
|
model integration time |
296 |
|
|
297 |
|
checkpoint51l_pre |
298 |
|
o add salt flux to forcing fields (used when salty sea-ice forms or melts) |
299 |
|
o enable to apply restoring on S & T only in a latitude band. |
300 |
|
o update output (external_forcing_surf modif. change truncation errors) |
301 |
|
global_ocean.cs32x15, global_ocean_pressure, lab_sea, natl_box |
302 |
|
|
303 |
|
checkpoint51k_post |
304 |
|
o change default to: useNHMTerms=.FALSE. |
305 |
|
- new output.txt for atmos. exp (NHMTerms were wrong but now turned off): |
306 |
|
adjustment.128x64x1, aim.5l_Eq_Ch, aim.5l_LatLon, hs94.128x64x5, hs94.1x64x5 |
307 |
|
- new output.txt for idealized oceanic exp (NHMTerms are now turned off): |
308 |
|
exp1, ideal_2D_oce |
309 |
|
- change data file of oceanic exp (leave the results unchanged for now) |
310 |
|
exp2, global_ocean[90x40x15,pressure,with_exf], lab_sea, natl_box |
311 |
|
o set gravitySign automatically according to the vertical coordinate used |
312 |
|
(and change sign: back to "logical" value: +1 when R increases downward) |
313 |
|
o added an include barrier to the default CPP_OPTIONS.h and |
314 |
|
removed the #include "PACKAGES_CONFIG.h" from it |
315 |
|
o fix a bug in mom_fluxform (from chk51j_post) |
316 |
|
|
317 |
|
checkpoint51j_post |
318 |
|
o Added some AD-related initialisations in mom_vecinv/ mom_fluxform/ |
319 |
|
o adjusted some flow directives (dynamics, thermodynamics) |
320 |
|
o added cubed-sphere case to test cost function |
321 |
|
o heckpoint_lev?_directives.h now include |
322 |
|
pkg-specific directives, whenever those are enabled. |
323 |
|
o Moving pkg-specific flow directives to corresponding pkg |
324 |
|
Moving pkg-specific differentiation list to corresponding pkg |
325 |
|
Moving pkg-specific checkpoint-lev storing to corresponding pkg |
326 |
|
---> conventions are |
327 |
|
- pkg_ad.flow |
328 |
|
- pkg_ad_diff.list |
329 |
|
- pkg_ad_check_lev?_dir.h |
330 |
|
|
331 |
|
checkpoint51i_post |
332 |
|
o merge of the branch-genmake2 |
333 |
|
- knowingly breaks the older genmake |
334 |
|
- automatic generation of all ALLOW_${PKG_NAME} defines |
335 |
|
- new testreport supersedes older testscript |
336 |
|
o large number of package inclusion and header cleanups |
337 |
|
checkpoint51i_pre |
338 |
|
|
339 |
|
o new output for adjoin of global ocean vector invariant |
340 |
|
o fix for checkpoint of thermodynamic seaice and bulk_force |
341 |
|
o Steph gchem various cleanups |
342 |
|
|
343 |
|
checkpoint 51h_pre |
344 |
|
o fixed some single quotes |
345 |
|
o modify gchem pkg to be more versatile |
346 |
|
o modifications to clean up dic pkg |
347 |
|
o add pkg/cfc |
348 |
|
|
349 |
|
checkpoint51g_post |
350 |
|
o provide links for ARPACK and ADM*TLM |
351 |
|
o start removing verification/vero/ |
352 |
|
o update AD-related stuff |
353 |
|
(in particular some IF-statements in model/src/ routines) |
354 |
|
o adding to verification/global_ocean.90x40x15/ adjoint of |
355 |
|
vector invariant setup |
356 |
|
|
357 |
|
checkpoint51f_post |
358 |
|
o update advect_xz/output.txt (left from chekpoint50f_pre, ini_vel modif) |
359 |
|
o Mods and bug fixes to pkg/cal, pkg/exf, etc., needed for computation |
360 |
|
of tracer Green's fucntions for ocean inversion project. |
361 |
|
|
362 |
|
checkpoint51f_pre |
363 |
|
o Added on-the-fly spatial interpolation capability |
364 |
|
"USE_EXF_INTERPOLATION" to pkg/exf. |
365 |
|
- This is a temporary Cartesian-grid hack until |
366 |
|
the super-duper ESMF coupler becomes available. |
367 |
|
- See verification/global_with_exf/README for usage example. |
368 |
|
- Removed obsolete EXFwindOnBgrid and SEAICEwindOnCgrid |
369 |
|
flags and modified pkg/seaice accordingly. |
370 |
|
o Bug fix to pkg/ptracers, pkg/generic_advdiff/gad_calc_rhs.F, |
371 |
|
and pkg/kpp/kpp_transport_ptr.F for dealing with tracer |
372 |
|
non-local transport term. |
373 |
|
|
374 |
|
checkpoint51e_post |
375 |
|
o pkg/mom_vecinv: |
376 |
|
- changes related to hFac: use S/R from aim.5l_cs/code (coriolis & vort.advec) |
377 |
|
- add time-ave diagnostic of vorticity advection. |
378 |
|
o add diagnostic (instantaneous) of Shapiro Filter effect for T,S & UV. |
379 |
|
o atmospheric geopotential: - include water vapor effect; |
380 |
|
- use p^kappa in p* scaling. |
381 |
|
o land pkg: read initial state from files. |
382 |
|
o AIM : add stratospheric drag in the upper level |
383 |
|
|
384 |
|
checkpoint51d_post |
385 |
|
o change function ILNBLNK (=> last non-blank char.) to return zero |
386 |
|
(instead of the string length) when the string. char is empty |
387 |
|
o fix bug in dxV,dyU definition of cubed-sphere grid (ini_curvilinear_grid) |
388 |
|
o modif's to restore TAMC compatibility |
389 |
|
- avoid "_d" constructs in PARAMETER declarations |
390 |
|
- avoid interrupting lines by comment lines |
391 |
|
o Use cluster local disks for purely local I/O |
392 |
|
vs. globally visible disks needed for ctrl stuff |
393 |
|
(merged from ecco-branch) |
394 |
|
- new active_file_loc... and mdsio_..._loc routines |
395 |
|
- affected packages: autodiff, ctrl, grdchk, mdsio |
396 |
|
|
397 |
|
checkpoint51c_post |
398 |
|
o introducing integer flag debugLevel |
399 |
|
- debLevZero: suppress all output |
400 |
|
- debLevA: standard |
401 |
|
- debLevB: equivalent to current debugMode=.TRUE. |
402 |
|
o introducing pathname variable mdsioLocalDir for mdsio |
403 |
|
that can be specified at runtime |
404 |
|
(relevant for cluster I/O to local disk) |
405 |
|
` |
406 |
|
>>>checkpoint51b_post |
407 |
|
adjoint of ptracers |
408 |
|
o disentangled ALLOW_PTRACERS using new ALLOW_GCHEM |
409 |
|
o extended state in checkpoint_lev?, etc. |
410 |
|
o alternative ctrl/cost with tr1 or ptracer |
411 |
|
o new comlev1_bibj_ptracers |
412 |
|
o new ptracers.flow |
413 |
|
o made ptracers_forcing same as external_forcing_s |
414 |
|
o added verification for adjoint with ptracers |
415 |
|
- adm. grdchk fail so far |
416 |
|
- tlm. grdchk OK |
417 |
|
- checked tapelev/comlev recomputations: OK |
418 |
|
<<<checkpoint51b_pre |
419 |
|
|
420 |
|
checkpoint51a_post |
421 |
|
o added code for biogeochemistry package |
422 |
|
- new pkgs: dic, gchem |
423 |
|
- new subroutine in ptracers |
424 |
|
o modify code for biogeo packages |
425 |
|
- in model/src and pkg/ptracers |
426 |
|
|
427 |
|
checkpoint51 |
428 |
|
o merged improved packages from ecco-branch (ecco_c51_e34) to MAIN |
429 |
|
- autodiff, ctrl, exf, grdchk, seaice |
430 |
|
kept separate packages |
431 |
|
- cost, ecco, obcs |
432 |
|
o updating verification/internal_wave, verification/lab_sea |
433 |
|
(all verif's pass on faulks.lcs.mit.edu RH7.3, g77) |
434 |
|
|
435 |
|
checkpoint50i_post |
436 |
|
o Preparing next differentiable checkpoint and sync |
437 |
|
of MAIN vs. ecco-branch |
438 |
|
- 1st step: this one is differentiable |
439 |
|
- 2nd step: next(! not this) checkpoint will have updated packages |
440 |
|
from ecco-branch) |
441 |
|
|
442 |
|
checkpoint50h_post |
443 |
|
o new package "land": provide surface BC for AIM. |
444 |
|
o pkg/ptracers: apply exchange to all tracers after reading a pickup. |
445 |
|
o AIM: use true log-P extrapolation for near surface temp (= as in |
446 |
|
F.Molteni paper) ; update aim.5l_cs output. |
447 |
|
o New variable in PARM03 for more flexible pickups |
448 |
|
o somewhat cleaned package initialisation sequence for |
449 |
|
ctrl/ cost/ ecco/ in packages_... routines |
450 |
|
|
451 |
|
checkpoint50g_post |
452 |
|
o merged with release1_p17 (pkg/seaice and verification/lab_sea) |
453 |
|
- added SEAICE_MULTILEVEL for 8-category sea-ice thermodynamics |
454 |
|
- LSR sea-ice dynamic solver moved to SouthWest B-grid location and |
455 |
|
made the default because of faster convergence than ADI |
456 |
|
|
457 |
|
checkpoint50f_post |
458 |
|
o exf totally restructured |
459 |
|
- replaced all exf_set_ routines by one generic exf_set_gen.F |
460 |
|
and made sure it's still differentiable |
461 |
|
- split exf_getffields.F into new exf_bulkformulae.F |
462 |
|
- exf_inscal_/exf_outscal_ suffixes consistent with field names |
463 |
|
- updated adjoint/makefile and relevant verif. accordingly |
464 |
|
|
465 |
|
checkpoint50f_pre |
466 |
|
o AIM: near surface temp evaluated from Pot.Temp: fixed for Part Cell ; |
467 |
|
+ change Evap over land. (update aim.5l_cs output) |
468 |
|
o added missing .cvsignores and made uniform across expts |
469 |
|
o changed configuration of tidal_basin_2d expt |
470 |
|
o added pkg/debug functions for tracing code in fatal crashes |
471 |
|
o tidied pkg/monitor to use MONITOR_OPTIONS.h |
472 |
|
o new fn in pkg/monitor to stop model if solution grows too big |
473 |
|
o removed lots of single quotes from comments |
474 |
|
|
475 |
|
checkpoint50e_post |
476 |
|
o make KPP work with PTRACERS |
477 |
|
- fix gad_calc_rhs to call new routine kpp_transport_ptr, which is |
478 |
|
nearly a copy of kpp_transport_s |
479 |
|
- there is no analogue to SurfaceTendencyS, so I have to use |
480 |
|
gPtr(of the surface layer) instead |
481 |
|
o add a new platform SunFire+mpi (SunFire 15000) to genmake |
482 |
|
checkpoint50e_pre |
483 |
|
|
484 |
|
checkpoint50d_post |
485 |
|
o change kpp output from multiple-record state files to single-record state |
486 |
|
files analogous to write_state.F |
487 |
|
o reduce the output frequency of cg3d-related stuff to the monitor frequency, |
488 |
|
analogous to the cg2d-related output. |
489 |
|
o fix small problem with in ptracers_write_checkpoint.F: len(suff)=512, |
490 |
|
so that writing to internal file fn (with length 512) fails. |
491 |
|
checkpoint50d_pre |
492 |
|
|
493 |
|
checkpoint50c_post |
494 |
|
o fix problem in ini_procs.F when using MPI (was there since checkpoint48f) |
495 |
|
o add monitor diagnostic of vorticity (mon_vort3.F) and potential energy. |
496 |
|
o add time-average diagnostic of U*V and vertical diffusive flux for theta. |
497 |
|
o bug fixed in convective adjustment diagnostic (multi-tile set-up). |
498 |
|
o 4th Order Adv scheme: return to previous (c40pre7) masking near boundaries. |
499 |
|
|
500 |
|
checkpoint50c_pre |
501 |
|
Merging from release1_p13: |
502 |
|
o bug fix for pkg/seaice dynamic solver |
503 |
|
o Added SEAICE_initialHEFF to pkg/seaice |
504 |
|
|
505 |
|
checkpoint50b_post |
506 |
|
o store u*,v* in gU,V instead of in guNm1,gvNm1 |
507 |
|
o in order to put the momForcing out of the Adams-Bashforth: |
508 |
|
move forcing & CD-scheme calls from mom_fluxform & mom_vecinv |
509 |
|
to timestep.F |
510 |
|
o new flag "useCDscheme" (default=F); replace guCD,gvCD by local arrays |
511 |
|
o re-generate output for highly sensitive test-exp: |
512 |
|
global_ocean_pressure, ideal_2D_oce, lab_sea, natl_box |
513 |
|
|
514 |
|
checkpoint50b_pre |
515 |
|
o restore the use of momAdvection & useCoriolis flags (mom_vecinv) |
516 |
|
o corrected CPP option mismatch in exf_Bulk... functions for case |
517 |
|
#define ALLOW_ATM_WIND && #undef ALLOW_ATM_TEMP |
518 |
|
|
519 |
|
checkpoint50a_post |
520 |
|
o Bug fix for merging between c50 and KPP. |
521 |
|
ikey was passed from thermodynamics to kpp_calc via |
522 |
|
common block rather than being recomputed in kpp_calc, |
523 |
|
in contradiction with new key itdkey. |
524 |
|
New key ikppkey created, and tamc.h headers updated. |
525 |
|
|
526 |
|
checkpoint50 |
527 |
|
- 2nd step of merge with ecco-branch: |
528 |
|
merge changes in ecco-branch back into main trunk |
529 |
|
- bug fix in find_alpha |
530 |
|
- more storing in gmredi_calc_tensor.F, kpp_routines.F |
531 |
|
- avoid iterative loop in ini_pressure for ALLOW_AUTODIFF_TAMC |
532 |
|
- merged some sliced AD support routines |
533 |
|
|
534 |
|
checkpoint49 |
535 |
|
merged from ecco-branch: |
536 |
|
o exf: |
537 |
|
- Enable initialisation of forcing fields to constant |
538 |
|
(runtime) values. |
539 |
|
- in exf_getffields.F |
540 |
|
Reduce i-/j-loop to interior domain, discarding overlaps. |
541 |
|
That also fixes wrong TAF-key computations for key_1, key_2 |
542 |
|
with bulf formulae. |
543 |
|
- exf_init.F modify #ifdef for exf_init_evap |
544 |
|
- exf_getffieldrec.F, ctrl_getrec.F |
545 |
|
The following INT-usages are not safe: |
546 |
|
fldsecs = int(fldsecs/fldperiod)*fldperiod |
547 |
|
fldcount = int(fldsecs/fldperiod) + 1 |
548 |
|
and were modified. |
549 |
|
o autodiff: checkpoint_lev... |
550 |
|
- updated storing for bulk formulae (some arrays were missing) |
551 |
|
- added state for seaice package |
552 |
|
- synchronized with main branch |
553 |
|
o DIVA: |
554 |
|
Modif.s in |
555 |
|
- the_main_loop.F |
556 |
|
- the_model_main.F |
557 |
|
and added cost_final_restore.F cost_final_store.F |
558 |
|
o With these changes, it is possible to fully separate |
559 |
|
the first forward run to compute the cost function (mdthe_main_loop) |
560 |
|
and adjoint runs over individual DIVA intervals (adthe_main_loop) |
561 |
|
into separate model executions. |
562 |
|
This enables better tuning of DIVA to given queue sizes. |
563 |
|
o Loss of cost function value in consecutive DIVA runs is |
564 |
|
fixed by storing to 'costfinal' file. |
565 |
|
o Appropriate call of ctrl_unpack/_pack,grdchk_main at |
566 |
|
beginning/end of adjoint integration is accounted for. |
567 |
|
The current changes apply to TAF-1.4.26. |
568 |
|
TAF has one remaining bug, requiring one manual change |
569 |
|
(see MITgcm manual). |
570 |
|
o cal: synchronized between MAIN and ecco-branch |
571 |
|
|
572 |
|
checkpoint48i_post |
573 |
|
o Changes to restore differentiability of code w.r.t. previous tag |
574 |
|
(mostly adding new routines to make list and replacing |
575 |
|
pressure by totPhiHyd). |
576 |
|
o Updated and merged grdchk package |
577 |
|
- has both ADM and TLM checks |
578 |
|
- works for single- and multi-proc. |
579 |
|
- output cleaned |
580 |
|
- worked successfully for parallel DIVA |
581 |
|
o genmake: |
582 |
|
- Adding -platform=SP4 |
583 |
|
- Modif. for DEC Alpha (many don't have makedepend; use mkdep instead). |
584 |
|
o Modif. for DIVA: |
585 |
|
ctrl_unpack should not be called in intermediate DIVA interval. |
586 |
|
(similar thing would apply for active files such as |
587 |
|
tbar, sbar, psbar, which should not be reinitialised; |
588 |
|
only relevant for ecco-branch) |
589 |
|
|
590 |
|
checkpoint48h_post |
591 |
|
o improve "ini_p_ground" when using finite difference form for calc_phi_hyd |
592 |
|
o Added net flux to downward flux conversion to pkg/exf/exf_getffields.F |
593 |
|
o Added SEAICE_initialHEFF to pkg/seaice |
594 |
|
o ecco_the_model_main.F: SEAICE_WRITE_PICKUP must precede WRITE_CHECKPOINT |
595 |
|
|
596 |
|
checkpoint48g_post |
597 |
|
--- EOS: |
598 |
|
o compute locally the pressure for use in EOS: UNESCO, JMD95P, MDJWF |
599 |
|
store total Potential in totPhyHyd for diagnostics & EOS. |
600 |
|
o fix restart and overlap Pb when using Z-coord and EOS funct. of P |
601 |
|
--- Pressure/geopotential gradient: |
602 |
|
o use Ro_surf & R_low instead of hFac in cal_phi_hyd: |
603 |
|
o NLFS + p-coord.: eta*Alpha' is not longer lagging 1.time-step behind |
604 |
|
( changes results of global_ocean_pressure ) |
605 |
|
o NLFS + z-coord.: add eta*Rho' contribution. |
606 |
|
o accurate phiHyd when using interface-W at the middle between 2 centers |
607 |
|
( changes results of ideal_2D_oce ) |
608 |
|
o includes r* 2nd term in potential gradient. |
609 |
|
|
610 |
|
checkpoint48f_post |
611 |
|
Merging from release1_p12: |
612 |
|
o Modifications for using pkg/exf with pkg/seaice |
613 |
|
- improved description of the various forcing configurations |
614 |
|
- added basic radiation bulk formulae to pkg/exf |
615 |
|
- units/sign fix for evap computation in exf_getffields.F |
616 |
|
- updated verification/global_with_exf/results/output.txt |
617 |
|
o Added pkg/sbo for computing IERS Special Bureau for the Oceans |
618 |
|
(SBO) core products, including oceanic mass, center-of-mass, |
619 |
|
angular, and bottom pressure (see pkg/sbo/README.sbo). |
620 |
|
o Lower bound for viscosity/diffusivity in pkg/kpp/kpp_routines.F |
621 |
|
to avoid negative values in shallow regions. |
622 |
|
- updated verification/natl_box/results/output.txt |
623 |
|
- updated verification/lab_sea/results/output.txt |
624 |
|
o MPI gather, scatter: eesupp/src/gather_2d.F and scatter_2d.F |
625 |
|
o Added useSingleCpuIO option (see PARAMS.h). |
626 |
|
o Updated useSingleCpuIO option in mdsio_writefield.F to |
627 |
|
work with multi-field files, e.g., for single-file pickup. |
628 |
|
o pkg/seaice: |
629 |
|
- bug fix in growth.F: QNET for no shortwave case |
630 |
|
- added HeffFile for specifying initial sea-ice thickness |
631 |
|
- changed SEAICE_EXTERNAL_FLUXES wind stress implementation |
632 |
|
o Added missing /* */ to CPP comments in pkg/seaice, pkg/exf, |
633 |
|
kpp_transport_t.F, forward_step.F, and the_main_loop.F |
634 |
|
o pkg/seaice: |
635 |
|
- adjoint-friendly modifications |
636 |
|
- added a SEAICE_WRITE_PICKUP at end of the_model_main.F |
637 |
|
|
638 |
|
checkpoint48e_post |
639 |
|
o New S/R for diagnostic of bottom pressure (phi0surf was missing in c48d_post) |
640 |
|
o dynamics: change definition of computational domain & adapt mom_fluxform |
641 |
|
accordingly ; when viscA4=0, allows to run the dynamics with Olx=Oly=2. |
642 |
|
o following exp works with Olx=Oly=2 (and pass through checkbounds option): |
643 |
|
adjust_nlfs.cs-32x32x1 adjustment.128x64x1 adjustment.cs-32x32x1 |
644 |
|
exp0 exp1 exp2 exp4 exp5 hs94.cs-32x32x5 hs94.128x64x5 hs94.1x64x5 |
645 |
|
global_ocean.90x40x15 global_with_exf global_ocean_pressure |
646 |
|
inverted_barometer solid-body.cs-32x32x1 |
647 |
|
(high order AdvScheme and/or multiDimAd works with Olx=Oly=2 ? not sure) |
648 |
|
|
649 |
|
checkpoint48d_post |
650 |
|
o in preparation for r*: |
651 |
|
a) use pre-computed gradient of hydrostatic potential: |
652 |
|
changes in timestep.F & mom_cdscheme.F affects results of ideal_2D_oce |
653 |
|
b) move phi0surf from calc_phi_hyd to calc_grad_phi_hyd: |
654 |
|
=> affects results of glob_oce_pressure (different truncation error) |
655 |
|
update output.txt of ideal_2D_oce & global_ocean_pressure |
656 |
|
|
657 |
|
checkpoint48d_pre |
658 |
|
o in preparation for r*: |
659 |
|
new S/R (calc_grad_phi_hyd.F) to compute Hydrostatic potential gradient. |
660 |
|
pass the 2 comp. of the grad. as arguments to momentum S/R. |
661 |
|
but for the moment, only used if it does not change the results. |
662 |
|
o add finite volume form of calc_phi_hyd for buoyancyRelation OCEANIC & OCEANICP |
663 |
|
|
664 |
|
checkpoint48c_post |
665 |
|
Modifications in treatment of initial flux forcing controls |
666 |
|
to avoid extra storing of flux fields. |
667 |
|
|
668 |
|
checkpoint48c_pre |
669 |
|
o Added tidal_basin_2d experiment (currently 1D!) |
670 |
|
o can use r* Coordinate with OBC (but Orlanski Rad. BC not yet implemented) |
671 |
|
|
672 |
|
checkpoint48b_post |
673 |
|
o r* coordinate added in #ifdef NONLIN_FRSURF block. |
674 |
|
(modification to pressure gradient not yet implemented) |
675 |
|
o mom_fluxform: new S/R mom_calc_rtrans.F computes vertical transports |
676 |
|
(for U & V) that are used for vertical advection of momentum. |
677 |
|
|
678 |
|
checkpoint48a_post |
679 |
|
o In preparation for r*, use temp. array to compute vertical transport |
680 |
|
in flux-form momentum pkg. => change results of few test-exp. |
681 |
|
Update output of: exp4, exp5, front_relax, global_ocean_pressure, |
682 |
|
lab_sea, natl_box & plume_on_slope |
683 |
|
|
684 |
|
checkpoint48 |
685 |
|
o fix small bug that was causing lot of warning (TIMER_STOP 'UPDATE_CG2D') |
686 |
|
|
687 |
|
checkpoint47j_post |
688 |
|
o add new exp. (ideal_2D_oce) to test "exotic" parameters and options. |
689 |
|
o moved adexch from forward_step to addummy_in_stepping |
690 |
|
o added exch in ctrl_map_... |
691 |
|
o aveFreq.=0. at end of cost_final to avoid overwrite in |
692 |
|
reverse checkpointing loop |
693 |
|
o Yet more changes in gmredi: |
694 |
|
* adgmredi_calc_tensor |
695 |
|
avoiding all recomputation of gmredi_slope_limit |
696 |
|
* adgmredi_x/y/rtransport |
697 |
|
added flag for excessive storing to avoid recomp. of |
698 |
|
u/v/rtans, dTdx/y/z |
699 |
|
-> this is not really necessary and very memory-consuming |
700 |
|
* adgmredi_slope_psi: |
701 |
|
consistency with gmredi_slope_limit in treatment of GM_slopeSqCutoff |
702 |
|
* gmredi_slope_limit |
703 |
|
re-activated full calculation of taperfct for case 'ac02' |
704 |
|
o updating verification/ adjoints of carbon, global_ocean.90x40x15 |
705 |
|
|
706 |
|
checkpoint47i_post |
707 |
|
o GM_EXCLUDE_() replace GM_TAPER_(): by default, all the code is compiled. |
708 |
|
o add GM_Small_Number & GM_slopeSqCutoff as run-time parameters (data.gmredi) |
709 |
|
default values = 1.D-12 & 1.D+48 (=> give same results as checkpoint47f_post) |
710 |
|
o cg2dTargetResWunit: define tolerance using deltaTmom (and not deltaTfreesurf) |
711 |
|
o gmredi: fix few bugs. |
712 |
|
|
713 |
|
checkpoint47h_post (duplicate tag for PH's c48 tag, in order to move c48) |
714 |
|
This will be a good checkpoint to plug into the ecco-branch. |
715 |
|
o few modif.'s to get latest version adjointed |
716 |
|
(mainly kick out code in ini_linear_phisurf) |
717 |
|
o modif's to run adjoint with exactConserv |
718 |
|
o case GM_BOLUS_ADVEC should be cleaned |
719 |
|
S/R gmredi_slope_psi should be cleaned |
720 |
|
o verification/carbon now has exactConserv=.TRUE. |
721 |
|
|
722 |
|
checkpoint47g_post |
723 |
|
o time-average AIM output: becomes independent from statvars time-average |
724 |
|
(can be used with #undef ALLOW_TIMEAVE) ; add new variables (e.g., RH) |
725 |
|
o After TAF bug fix: |
726 |
|
Tested adjoint of gmredi package. |
727 |
|
Tests were done with checkpoint47 rather than with latest |
728 |
|
checkpoint47f_post since the latter is broken for the adjoint |
729 |
|
verification/carbon/ contains both ADM and TLM gradient checks. |
730 |
|
Here they are, before they get changed/lost/stolen. |
731 |
|
Mostly modif.'s to fix numerical sensitivities. |
732 |
|
Gradient checks OK for |
733 |
|
- GM_taper_scheme: |
734 |
|
* clipping |
735 |
|
* ac02 |
736 |
|
* linear |
737 |
|
* glw91 |
738 |
|
* dm95 |
739 |
|
* ldd97 |
740 |
|
- GMREDI_OPTIONS: |
741 |
|
* GM_VISBECK_VARIABLE_K |
742 |
|
* GM_NON_UNITY_DIAGONAL |
743 |
|
* GM_EXTRA_DIAGONAL |
744 |
|
* GM_BOLUS_ADVEC |
745 |
|
in conjunction with data.gmredi parameters to be checked in |
746 |
|
in a few minutes under verification/carbon/code/ |
747 |
|
|
748 |
|
checkpoint47f_post |
749 |
|
Merging from release1_p10: |
750 |
|
o modifications for using pkg/exf with pkg/seaice |
751 |
|
- pkg/seaice CPP options SEAICE_EXTERNAL_FORCING |
752 |
|
and SEAICE_EXTERNAL_FLUXES |
753 |
|
- pkg/exf CPP options EXF_READ_EVAP and |
754 |
|
EXF_NO_BULK_COMPUTATIONS |
755 |
|
- usage examples are Experiments 8 and 9 in |
756 |
|
verification/lab_sea/README |
757 |
|
- verification/lab_sea default experiment now uses |
758 |
|
pkg/gmredi, pkg/kpp, pkg/seaice, and pkg/exf |
759 |
|
|
760 |
|
checkpoint47e_post |
761 |
|
o created a tag prior to creating a branch |
762 |
|
|
763 |
|
checkpoint47d_post |
764 |
|
o fixed bug in the oceanic pressure coordinates code: vertical viscosity |
765 |
|
at the bottom boundary had an erroneous half slip boundary condition |
766 |
|
o added bottom drag and no slip boundary condition capabilities to |
767 |
|
oceanic pressure coordinates code |
768 |
|
o added new verification experiment global_ocean_pressure, this experiment is |
769 |
|
also described in the tutorial section of the manual |
770 |
|
checkpoint47d_pre |
771 |
|
|
772 |
|
checkpoint47c_post |
773 |
|
o Added a verification experiment for a global cubed-sphere ocean run |
774 |
|
- verification/global_ocean.cubed32x32x30 |
775 |
|
- uses pkg/therm_seaice and pkg/bulk_forcing |
776 |
|
|
777 |
|
o allow to run AIM physics with SPEEDY input files (from Franco Molteni) |
778 |
|
o allow a more accurate definition of Ro_Surf (selectFindRoSurf=1) |
779 |
|
when using P-coordinate; only implemented for atmospheric config. |
780 |
|
o OCEANICP & realFreshWater: include P-E direct effect on wVel ; |
781 |
|
NOTES: requires option NONLIN_FRSURF to be "#define". |
782 |
|
o update advect_xz/results/output.txt (left from checkpoint44g_post) |
783 |
|
|
784 |
|
checkpoint47b_post |
785 |
|
Merging from release1_p9: |
786 |
|
o pkg/seaice |
787 |
|
- removed GOTO's and added taf directives |
788 |
|
- double precision constants to reduce the g77 (Linux) |
789 |
|
to F77 (SGI) differences reported in release1_p8 |
790 |
|
o tools/genmake |
791 |
|
- added SGI options |
792 |
|
o verification/testscript |
793 |
|
- updated to that of checkpoint47a_post |
794 |
|
o verification/global_ocean.90x40x15/input/eedata |
795 |
|
- modified for SGI f77 compatibility |
796 |
|
o verification/lab_sea |
797 |
|
- added description of sea-ice model |
798 |
|
- added missing matlab routines |
799 |
|
- added test of thermodynamics parallelization |
800 |
|
|
801 |
|
checkpoint47a_post |
802 |
|
o new pkg: aim_v23 |
803 |
|
= F. Molteni atmos.physics (SPEEDY, ver23) adapted to MITgcm. |
804 |
|
for now, keep the same surface forcing as before. |
805 |
|
- Part-Cell implemented into AIM; check that Heat & Water are conserved |
806 |
|
- aim.5l_cs: use new aim pkg (run 10yr & get better results than before) |
807 |
|
- aim.5l_LatLon & aim.5l_Equatorial_Channel are still using old aim pkg |
808 |
|
o add diagnostic of surface correction term in monitor output |
809 |
|
|
810 |
|
o bulk_force and therm_seaice |
811 |
|
- Two packages: bulk_force (Bulk forcing) |
812 |
|
and therm_seaice (thermodynamic_seaice) - adopted from LANL CICE.v2.0.2 |
813 |
|
- Earlier integration from Stephaine Dutkiewicz |
814 |
|
and Patrick Heimbach. |
815 |
|
- Two ifdef statements for compile time, |
816 |
|
ALLOW_THERM_SEAICE and ALLOW_BULK_FORCE |
817 |
|
|
818 |
|
checkpoint47 |
819 |
|
differentiable version of checkpoint46n_post |
820 |
|
o bug fix in quasihydrostaticterms |
821 |
|
o * "clean" adjoint code (in terms of extensive recomputations) |
822 |
|
can now be obtained for all GMREDI options (i.e. for |
823 |
|
- GM_VISBECK_VARIABLE_K |
824 |
|
- GM_NON_UNITY_DIAGONAL |
825 |
|
- GM_EXTRA_DIAGONAL |
826 |
|
- GM_BOLUS_ADVEC ) |
827 |
|
* However, wrong gradient check problem remains unsolved. |
828 |
|
* New CPP options have been introduced for different |
829 |
|
tapering schemes |
830 |
|
o external_fields_load now part of differentiation list |
831 |
|
o pressure needs multiple storing; |
832 |
|
would be nice to have store_pressure at beginning or |
833 |
|
end of forward_step, e.g. by having phiHyd global (5-dim.) |
834 |
|
(NB: pressure is needed for certain cases in find_rho, |
835 |
|
which is also invoked through convective_adjustment). |
836 |
|
o recomputations in find_rho for cases |
837 |
|
'JMD95'/'UNESCO' or 'MDJWF' are OK. |
838 |
|
o #define ATMOSPHERIC_LOADING should be differentiable |
839 |
|
o ini_forcing shifted to begining of initialise_varia |
840 |
|
o Incorporating QNVS line search routines into MITgcm |
841 |
|
(this is separate code, and therefore not under pkg) |
842 |
|
lsopt/, optim/ |
843 |
|
o Updated verification/carbon/ |
844 |
|
- #define GM_VISBECK_VARIABLE_K |
845 |
|
- GM_taper_scheme = 'ldd97' |
846 |
|
|
847 |
|
checkpoint46n_post |
848 |
|
Merging from release1_p8: |
849 |
|
o verification/natl_box: |
850 |
|
updating new external_fields_load routine |
851 |
|
o New package: pkg/seaice |
852 |
|
Sea ice model by D. Menemenlis (JPL) and Jinlun Zhang (Seattle). |
853 |
|
The sea-ice code is based on Hibler (1979-1980). |
854 |
|
Two sea-ice dynamic solvers, ADI and LSR, are included. |
855 |
|
In addition to computing prognostic sea-ice variables and diagnosing |
856 |
|
the forcing/external data fields that drive the ocean model, |
857 |
|
SEAICE_MODEL also sets theta to the freezing point under sea-ice. |
858 |
|
The implied surface heat flux is then stored in variable |
859 |
|
surfaceTendencyTice, which is needed by KPP package (kpp_calc.F and |
860 |
|
kpp_transport_t.F) to diagnose surface buoyancy fluxes and for the |
861 |
|
non-local transport term. Because this call precedes model |
862 |
|
thermodynamics, temperature under sea-ice may not be "exactly" at |
863 |
|
the freezing point by the time theta is dumped or time-averaged. |
864 |
|
N.B.: fairly large differences in accuracy occur across |
865 |
|
different platforms/compilers; comparison between |
866 |
|
g77 (Linux) and F77 (SGI Origin 2000/MIPSpro) |
867 |
|
generated output gives: |
868 |
|
T S U V |
869 |
|
C D M c m s m s m s m s |
870 |
|
n p a R g m m e . m m e . m m e . m m e . |
871 |
|
f n k u 2 i a a d i a a d i a a d i a a d |
872 |
|
g d e n d n x n . n x n . n x n . n x n . |
873 |
|
|
874 |
|
Y Y Y Y 5 5 7 7 7 8 10 9 6 6 6 6 7 5 7 5 7 FAIL lab_sea |
875 |
|
o GAD: |
876 |
|
- generated new common blocks to account for call of |
877 |
|
same gad routines with differing traceridentities |
878 |
|
(needed to modify tracerIdentity indices in GAD.h) |
879 |
|
- generated separate common blocks for case useCubedSphereExchange |
880 |
|
(Department of Futurology) |
881 |
|
- parameter lists to gmredi_?transport: added tracerIdentity |
882 |
|
- added new key indices to tamc.h |
883 |
|
o external_fields_load: |
884 |
|
- added this routine to TAF list |
885 |
|
- needed to make some common blocks global and additional storing |
886 |
|
along the same lines as exf package (checkpoint_lev?_directives.h) |
887 |
|
o exf: |
888 |
|
updated external forcing package |
889 |
|
- bug fixes carried over from ecco-branch |
890 |
|
(missing OBCS_OPTIONS.h in two routines) |
891 |
|
- enable easy to use "no forcing". |
892 |
|
- added exf I/O for atmospheric loading |
893 |
|
- added exf I/O for runoff data |
894 |
|
- transfered scaling between exf <-> MITgcm to exf namelist |
895 |
|
o Adding new verification experiment global_with_exf. |
896 |
|
This verification experiment is almost identical to the |
897 |
|
global_ocean.90x40x15 experiment, except that it uses |
898 |
|
the exf (and cal) package instead of the |
899 |
|
external_fields_load routine (cf README file). |
900 |
|
|
901 |
|
checkpoint46m_post |
902 |
|
o modified testscript to return -- when there are no lines of "%MON" output |
903 |
|
o added new flag "quasiHydrostatic" for QH mode |
904 |
|
and new routine quasihydrostaticterms() |
905 |
|
o added 2*Omega*Cos(phi)*W to u equations (non-hydrostatic Coriolis term) |
906 |
|
New routine: mom_u_coriolis_nh() |
907 |
|
o fixed sign of non-hydrostatic metric terms in u and v equations. |
908 |
|
Depend on correct sign of gravitySign. |
909 |
|
o added work.pc* to .cvsignore in aim directories using ifc |
910 |
|
o introduced flag for controllin non-hydrostatic metric terms |
911 |
|
o removed gravitySign from PARM04 |
912 |
|
o switched to g77/ifc from pgf77 in aim experiments |
913 |
|
|
914 |
|
checkpoint46l_post |
915 |
|
o made convective adjustment work with pressure coordinates: |
916 |
|
- changed the direction of k-loop in convective_adjustment.F for the |
917 |
|
case of pressure coordinates (OCEANICP,ATMOSPHERIC buoyancyRelation) |
918 |
|
- adjusted the reference pressure k-index in convective_adjustment.F |
919 |
|
- adjusted the convection condition in convect.F (in analogy to |
920 |
|
calc_ivdc.F) |
921 |
|
- convective_adjustment no longer computes anything on the halos |
922 |
|
- removed the warnings about negative salinity from find_rho.F and |
923 |
|
find_alpha.F; instead the new routine look_for_neg_salinity, called |
924 |
|
at the beginning of find_rho, find_alpha, and find_beta, does a |
925 |
|
check of the entire slice, if CPP-option |
926 |
|
CHECK_SALINITY_FOR_NEGATIVE_VALUES is defined |
927 |
|
checkpoint46l_pre |
928 |
|
|
929 |
|
checkpoint46k_post |
930 |
|
o fixed the verification/global_ocean.90x40x15 experiment: |
931 |
|
- new bathymetry (the world according to A., JMC, and M.) |
932 |
|
- new initial fields and forcing fields (*.bin files) |
933 |
|
- new POLY3.COEFFS (for the next release one should switch to a full |
934 |
|
equation of state: JMD95P or MDJWF) |
935 |
|
- fixed several errors and redundancies in the data file |
936 |
|
- experiment uses looped cells |
937 |
|
- added matlab directory with diagnostic scripts for plotting of output |
938 |
|
|
939 |
|
o S/R aim_initialise.F replace S/R aim_init from file aim_do_inphys.F: |
940 |
|
- read AIM physics parameters from a file (data.aimphys) |
941 |
|
- set defaults values = F.Molteni paper (Clim.Dyn., 2002) |
942 |
|
|
943 |
|
checkpoint46j_post |
944 |
|
o split calc_exact_eta in 2 S/R : integr_continuity & update_etaH |
945 |
|
o move wVel computation at the end of the time step, in S/R integr_continuity |
946 |
|
o create specific S/R to exchange T,S before DYNAMICS (for stagger time step) |
947 |
|
o update timeave pkg for wVel diagnostic. |
948 |
|
checkpoint46j_pre |
949 |
|
|
950 |
|
checkpoint46i_post |
951 |
|
o Clean up AIM package (and keep the results unchanged): |
952 |
|
include CPP_OPTION and use IMPLICT NONE in all routines ; |
953 |
|
declare all the variables _RL ; use _d 0 for all numerical constants. |
954 |
|
use ifdef ALLOW_AIM everywhere. And now AIM can be used with g77 ! |
955 |
|
|
956 |
|
checkpoint46h_post |
957 |
|
o cleaned up the use of rhoNil and rhoConst. |
958 |
|
- rhoNil should only appear in the LINEAR equation of state, everywhere |
959 |
|
else rhoNil is replaced by rhoConst, e.g. find_rho computes rho-rhoConst |
960 |
|
and the dynamical equations are all divided by rhoConst |
961 |
|
o introduced new parameter rhoConstFresh, a reference density of fresh |
962 |
|
water, to remove the fresh water flux's dependence on rhoNil. The default |
963 |
|
value is 999.8 kg/m^3 |
964 |
|
o cleanup up external_forcing.F and external_forcing_surf.F |
965 |
|
- can now be used by both OCEANIC and OCEANICP |
966 |
|
checkpoint46h_pre |
967 |
|
|
968 |
|
Added code to convert surface volume flux (fresh water) into |
969 |
|
a mass flux when using P coordinates in the ocean (OCEANICP). |
970 |
|
Note: It assumes you have set rho0=rhoConst=density of fresh water. |
971 |
|
|
972 |
|
checkpoint46g_post |
973 |
|
o Include a new diagnostic variable phiHydLow for the ocean model |
974 |
|
- in z-coordinates, it is the bottom pressure anomaly |
975 |
|
- in p-coordinates, it is the sea surface elevation |
976 |
|
- in both cases, these variable have global drift, reflecting the mass |
977 |
|
drift in z-coordinates and the volume drift in p-coordinates |
978 |
|
- included time averaging for phiHydLow, be aware of the drift! |
979 |
|
o depth-dependent computation of Bo_surf for pressure coordinates |
980 |
|
in the ocean (buoyancyRelation='OCEANICP') |
981 |
|
- requires a new routine (FIND_RHO_SCALAR) to compute density with only |
982 |
|
Theta, Salinity, and Pressure in the parameter list. This routine is |
983 |
|
presently contained in find_rho.F. This routine does not give the |
984 |
|
correct density for 'POLY3', which would be a z-dependent reference |
985 |
|
density. |
986 |
|
o cleaned up find_rho |
987 |
|
- removed obsolete 'eqn' from the parameter list. |
988 |
|
o added two new verification experiments: gop and goz |
989 |
|
(4x4 degree global ocean, 15 layers in pressure and height coordinates) |
990 |
|
checkpoint46g_pre |
991 |
|
|
992 |
checkpoint46f_post |
checkpoint46f_post |
993 |
o Enable tangent linear (forward mode) gradient checks: |
o Enable tangent linear (forward mode) gradient checks: |