| 3 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
| 4 |
============================== |
============================== |
| 5 |
|
|
| 6 |
|
o can use r* Coordinate with OBC (but Orlanski Rad. BC not yet implemented) |
| 7 |
|
|
| 8 |
|
checkpoint48b_post |
| 9 |
|
o r* coordinate added in #ifdef NONLIN_FRSURF block. |
| 10 |
|
(modification to pressure gradient not yet implemented) |
| 11 |
|
o mom_fluxform: new S/R mom_calc_rtrans.F computes vertical transports |
| 12 |
|
(for U & V) that are used for vertical advection of momentum. |
| 13 |
|
|
| 14 |
|
checkpoint48a_post |
| 15 |
|
o In preparation for r*, use temp. array to compute vertical transport |
| 16 |
|
in flux-form momentum pkg. => change results of few test-exp. |
| 17 |
|
Update output of: exp4, exp5, front_relax, global_ocean_pressure, |
| 18 |
|
lab_sea, natl_box & plume_on_slope |
| 19 |
|
|
| 20 |
|
checkpoint48 |
| 21 |
|
o fix small bug that was causing lot of warning (TIMER_STOP 'UPDATE_CG2D') |
| 22 |
|
|
| 23 |
|
checkpoint47j_post |
| 24 |
|
o add new exp. (ideal_2D_oce) to test "exotic" parameters and options. |
| 25 |
|
o moved adexch from forward_step to addummy_in_stepping |
| 26 |
|
o added exch in ctrl_map_... |
| 27 |
|
o aveFreq.=0. at end of cost_final to avoid overwrite in |
| 28 |
|
reverse checkpointing loop |
| 29 |
|
o Yet more changes in gmredi: |
| 30 |
|
* adgmredi_calc_tensor |
| 31 |
|
avoiding all recomputation of gmredi_slope_limit |
| 32 |
|
* adgmredi_x/y/rtransport |
| 33 |
|
added flag for excessive storing to avoid recomp. of |
| 34 |
|
u/v/rtans, dTdx/y/z |
| 35 |
|
-> this is not really necessary and very memory-consuming |
| 36 |
|
* adgmredi_slope_psi: |
| 37 |
|
consistency with gmredi_slope_limit in treatment of GM_slopeSqCutoff |
| 38 |
|
* gmredi_slope_limit |
| 39 |
|
re-activated full calculation of taperfct for case 'ac02' |
| 40 |
|
o updating verification/ adjoints of carbon, global_ocean.90x40x15 |
| 41 |
|
|
| 42 |
|
checkpoint47i_post |
| 43 |
|
o GM_EXCLUDE_() replace GM_TAPER_(): by default, all the code is compiled. |
| 44 |
|
o add GM_Small_Number & GM_slopeSqCutoff as run-time parameters (data.gmredi) |
| 45 |
|
default values = 1.D-12 & 1.D+48 (=> give same results as checkpoint47f_post) |
| 46 |
|
o cg2dTargetResWunit: define tolerance using deltaTmom (and not deltaTfreesurf) |
| 47 |
|
o gmredi: fix few bugs. |
| 48 |
|
|
| 49 |
|
checkpoint47h_post (duplicate tag for PH's c48 tag, in order to move c48) |
| 50 |
|
This will be a good checkpoint to plug into the ecco-branch. |
| 51 |
|
o few modif.'s to get latest version adjointed |
| 52 |
|
(mainly kick out code in ini_linear_phisurf) |
| 53 |
|
o modif's to run adjoint with exactConserv |
| 54 |
|
o case GM_BOLUS_ADVEC should be cleaned |
| 55 |
|
S/R gmredi_slope_psi should be cleaned |
| 56 |
|
o verification/carbon now has exactConserv=.TRUE. |
| 57 |
|
|
| 58 |
|
checkpoint47g_post |
| 59 |
|
o time-average AIM output: becomes independent from statvars time-average |
| 60 |
|
(can be used with #undef ALLOW_TIMEAVE) ; add new variables (e.g., RH) |
| 61 |
|
o After TAF bug fix: |
| 62 |
|
Tested adjoint of gmredi package. |
| 63 |
|
Tests were done with checkpoint47 rather than with latest |
| 64 |
|
checkpoint47f_post since the latter is broken for the adjoint |
| 65 |
|
verification/carbon/ contains both ADM and TLM gradient checks. |
| 66 |
|
Here they are, before they get changed/lost/stolen. |
| 67 |
|
Mostly modif.'s to fix numerical sensitivities. |
| 68 |
|
Gradient checks OK for |
| 69 |
|
- GM_taper_scheme: |
| 70 |
|
* clipping |
| 71 |
|
* ac02 |
| 72 |
|
* linear |
| 73 |
|
* glw91 |
| 74 |
|
* dm95 |
| 75 |
|
* ldd97 |
| 76 |
|
- GMREDI_OPTIONS: |
| 77 |
|
* GM_VISBECK_VARIABLE_K |
| 78 |
|
* GM_NON_UNITY_DIAGONAL |
| 79 |
|
* GM_EXTRA_DIAGONAL |
| 80 |
|
* GM_BOLUS_ADVEC |
| 81 |
|
in conjunction with data.gmredi parameters to be checked in |
| 82 |
|
in a few minutes under verification/carbon/code/ |
| 83 |
|
|
| 84 |
|
checkpoint47f_post |
| 85 |
|
Merging from release1_p10: |
| 86 |
|
o modifications for using pkg/exf with pkg/seaice |
| 87 |
|
- pkg/seaice CPP options SEAICE_EXTERNAL_FORCING |
| 88 |
|
and SEAICE_EXTERNAL_FLUXES |
| 89 |
|
- pkg/exf CPP options EXF_READ_EVAP and |
| 90 |
|
EXF_NO_BULK_COMPUTATIONS |
| 91 |
|
- usage examples are Experiments 8 and 9 in |
| 92 |
|
verification/lab_sea/README |
| 93 |
|
- verification/lab_sea default experiment now uses |
| 94 |
|
pkg/gmredi, pkg/kpp, pkg/seaice, and pkg/exf |
| 95 |
|
|
| 96 |
|
checkpoint47e_post |
| 97 |
|
o created a tag prior to creating a branch |
| 98 |
|
|
| 99 |
|
checkpoint47d_post |
| 100 |
|
o fixed bug in the oceanic pressure coordinates code: vertical viscosity |
| 101 |
|
at the bottom boundary had an erroneous half slip boundary condition |
| 102 |
|
o added bottom drag and no slip boundary condition capabilities to |
| 103 |
|
oceanic pressure coordinates code |
| 104 |
|
o added new verification experiment global_ocean_pressure, this experiment is |
| 105 |
|
also described in the tutorial section of the manual |
| 106 |
|
checkpoint47d_pre |
| 107 |
|
|
| 108 |
|
checkpoint47c_post |
| 109 |
|
o Added a verification experiment for a global cubed-sphere ocean run |
| 110 |
|
- verification/global_ocean.cubed32x32x30 |
| 111 |
|
- uses pkg/therm_seaice and pkg/bulk_forcing |
| 112 |
|
|
| 113 |
|
o allow to run AIM physics with SPEEDY input files (from Franco Molteni) |
| 114 |
|
o allow a more accurate definition of Ro_Surf (selectFindRoSurf=1) |
| 115 |
|
when using P-coordinate; only implemented for atmospheric config. |
| 116 |
|
o OCEANICP & realFreshWater: include P-E direct effect on wVel ; |
| 117 |
|
NOTES: requires option NONLIN_FRSURF to be "#define". |
| 118 |
|
o update advect_xz/results/output.txt (left from checkpoint44g_post) |
| 119 |
|
|
| 120 |
|
checkpoint47b_post |
| 121 |
|
Merging from release1_p9: |
| 122 |
|
o pkg/seaice |
| 123 |
|
- removed GOTO's and added taf directives |
| 124 |
|
- double precision constants to reduce the g77 (Linux) |
| 125 |
|
to F77 (SGI) differences reported in release1_p8 |
| 126 |
|
o tools/genmake |
| 127 |
|
- added SGI options |
| 128 |
|
o verification/testscript |
| 129 |
|
- updated to that of checkpoint47a_post |
| 130 |
|
o verification/global_ocean.90x40x15/input/eedata |
| 131 |
|
- modified for SGI f77 compatibility |
| 132 |
|
o verification/lab_sea |
| 133 |
|
- added description of sea-ice model |
| 134 |
|
- added missing matlab routines |
| 135 |
|
- added test of thermodynamics parallelization |
| 136 |
|
|
| 137 |
|
checkpoint47a_post |
| 138 |
|
o new pkg: aim_v23 |
| 139 |
|
= F. Molteni atmos.physics (SPEEDY, ver23) adapted to MITgcm. |
| 140 |
|
for now, keep the same surface forcing as before. |
| 141 |
|
- Part-Cell implemented into AIM; check that Heat & Water are conserved |
| 142 |
|
- aim.5l_cs: use new aim pkg (run 10yr & get better results than before) |
| 143 |
|
- aim.5l_LatLon & aim.5l_Equatorial_Channel are still using old aim pkg |
| 144 |
|
o add diagnostic of surface correction term in monitor output |
| 145 |
|
|
| 146 |
|
o bulk_force and therm_seaice |
| 147 |
|
- Two packages: bulk_force (Bulk forcing) |
| 148 |
|
and therm_seaice (thermodynamic_seaice) - adopted from LANL CICE.v2.0.2 |
| 149 |
|
- Earlier integration from Stephaine Dutkiewicz |
| 150 |
|
and Patrick Heimbach. |
| 151 |
|
- Two ifdef statements for compile time, |
| 152 |
|
ALLOW_THERM_SEAICE and ALLOW_BULK_FORCE |
| 153 |
|
|
| 154 |
|
checkpoint47 |
| 155 |
|
differentiable version of checkpoint46n_post |
| 156 |
|
o bug fix in quasihydrostaticterms |
| 157 |
|
o * "clean" adjoint code (in terms of extensive recomputations) |
| 158 |
|
can now be obtained for all GMREDI options (i.e. for |
| 159 |
|
- GM_VISBECK_VARIABLE_K |
| 160 |
|
- GM_NON_UNITY_DIAGONAL |
| 161 |
|
- GM_EXTRA_DIAGONAL |
| 162 |
|
- GM_BOLUS_ADVEC ) |
| 163 |
|
* However, wrong gradient check problem remains unsolved. |
| 164 |
|
* New CPP options have been introduced for different |
| 165 |
|
tapering schemes |
| 166 |
|
o external_fields_load now part of differentiation list |
| 167 |
|
o pressure needs multiple storing; |
| 168 |
|
would be nice to have store_pressure at beginning or |
| 169 |
|
end of forward_step, e.g. by having phiHyd global (5-dim.) |
| 170 |
|
(NB: pressure is needed for certain cases in find_rho, |
| 171 |
|
which is also invoked through convective_adjustment). |
| 172 |
|
o recomputations in find_rho for cases |
| 173 |
|
'JMD95'/'UNESCO' or 'MDJWF' are OK. |
| 174 |
|
o #define ATMOSPHERIC_LOADING should be differentiable |
| 175 |
|
o ini_forcing shifted to begining of initialise_varia |
| 176 |
|
o Incorporating QNVS line search routines into MITgcm |
| 177 |
|
(this is separate code, and therefore not under pkg) |
| 178 |
|
lsopt/, optim/ |
| 179 |
|
o Updated verification/carbon/ |
| 180 |
|
- #define GM_VISBECK_VARIABLE_K |
| 181 |
|
- GM_taper_scheme = 'ldd97' |
| 182 |
|
|
| 183 |
|
checkpoint46n_post |
| 184 |
|
Merging from release1_p8: |
| 185 |
|
o verification/natl_box: |
| 186 |
|
updating new external_fields_load routine |
| 187 |
|
o New package: pkg/seaice |
| 188 |
|
Sea ice model by D. Menemenlis (JPL) and Jinlun Zhang (Seattle). |
| 189 |
|
The sea-ice code is based on Hibler (1979-1980). |
| 190 |
|
Two sea-ice dynamic solvers, ADI and LSR, are included. |
| 191 |
|
In addition to computing prognostic sea-ice variables and diagnosing |
| 192 |
|
the forcing/external data fields that drive the ocean model, |
| 193 |
|
SEAICE_MODEL also sets theta to the freezing point under sea-ice. |
| 194 |
|
The implied surface heat flux is then stored in variable |
| 195 |
|
surfaceTendencyTice, which is needed by KPP package (kpp_calc.F and |
| 196 |
|
kpp_transport_t.F) to diagnose surface buoyancy fluxes and for the |
| 197 |
|
non-local transport term. Because this call precedes model |
| 198 |
|
thermodynamics, temperature under sea-ice may not be "exactly" at |
| 199 |
|
the freezing point by the time theta is dumped or time-averaged. |
| 200 |
|
N.B.: fairly large differences in accuracy occur across |
| 201 |
|
different platforms/compilers; comparison between |
| 202 |
|
g77 (Linux) and F77 (SGI Origin 2000/MIPSpro) |
| 203 |
|
generated output gives: |
| 204 |
|
T S U V |
| 205 |
|
C D M c m s m s m s m s |
| 206 |
|
n p a R g m m e . m m e . m m e . m m e . |
| 207 |
|
f n k u 2 i a a d i a a d i a a d i a a d |
| 208 |
|
g d e n d n x n . n x n . n x n . n x n . |
| 209 |
|
|
| 210 |
|
Y Y Y Y 5 5 7 7 7 8 10 9 6 6 6 6 7 5 7 5 7 FAIL lab_sea |
| 211 |
|
o GAD: |
| 212 |
|
- generated new common blocks to account for call of |
| 213 |
|
same gad routines with differing traceridentities |
| 214 |
|
(needed to modify tracerIdentity indices in GAD.h) |
| 215 |
|
- generated separate common blocks for case useCubedSphereExchange |
| 216 |
|
(Department of Futurology) |
| 217 |
|
- parameter lists to gmredi_?transport: added tracerIdentity |
| 218 |
|
- added new key indices to tamc.h |
| 219 |
|
o external_fields_load: |
| 220 |
|
- added this routine to TAF list |
| 221 |
|
- needed to make some common blocks global and additional storing |
| 222 |
|
along the same lines as exf package (checkpoint_lev?_directives.h) |
| 223 |
|
o exf: |
| 224 |
|
updated external forcing package |
| 225 |
|
- bug fixes carried over from ecco-branch |
| 226 |
|
(missing OBCS_OPTIONS.h in two routines) |
| 227 |
|
- enable easy to use "no forcing". |
| 228 |
|
- added exf I/O for atmospheric loading |
| 229 |
|
- added exf I/O for runoff data |
| 230 |
|
- transfered scaling between exf <-> MITgcm to exf namelist |
| 231 |
|
o Adding new verification experiment global_with_exf. |
| 232 |
|
This verification experiment is almost identical to the |
| 233 |
|
global_ocean.90x40x15 experiment, except that it uses |
| 234 |
|
the exf (and cal) package instead of the |
| 235 |
|
external_fields_load routine (cf README file). |
| 236 |
|
|
| 237 |
|
checkpoint46m_post |
| 238 |
|
o modified testscript to return -- when there are no lines of "%MON" output |
| 239 |
|
o added new flag "quasiHydrostatic" for QH mode |
| 240 |
|
and new routine quasihydrostaticterms() |
| 241 |
|
o added 2*Omega*Cos(phi)*W to u equations (non-hydrostatic Coriolis term) |
| 242 |
|
New routine: mom_u_coriolis_nh() |
| 243 |
|
o fixed sign of non-hydrostatic metric terms in u and v equations. |
| 244 |
|
Depend on correct sign of gravitySign. |
| 245 |
|
o added work.pc* to .cvsignore in aim directories using ifc |
| 246 |
|
o introduced flag for controllin non-hydrostatic metric terms |
| 247 |
|
o removed gravitySign from PARM04 |
| 248 |
|
o switched to g77/ifc from pgf77 in aim experiments |
| 249 |
|
|
| 250 |
|
checkpoint46l_post |
| 251 |
|
o made convective adjustment work with pressure coordinates: |
| 252 |
|
- changed the direction of k-loop in convective_adjustment.F for the |
| 253 |
|
case of pressure coordinates (OCEANICP,ATMOSPHERIC buoyancyRelation) |
| 254 |
|
- adjusted the reference pressure k-index in convective_adjustment.F |
| 255 |
|
- adjusted the convection condition in convect.F (in analogy to |
| 256 |
|
calc_ivdc.F) |
| 257 |
|
- convective_adjustment no longer computes anything on the halos |
| 258 |
|
- removed the warnings about negative salinity from find_rho.F and |
| 259 |
|
find_alpha.F; instead the new routine look_for_neg_salinity, called |
| 260 |
|
at the beginning of find_rho, find_alpha, and find_beta, does a |
| 261 |
|
check of the entire slice, if CPP-option |
| 262 |
|
CHECK_SALINITY_FOR_NEGATIVE_VALUES is defined |
| 263 |
|
checkpoint46l_pre |
| 264 |
|
|
| 265 |
|
checkpoint46k_post |
| 266 |
|
o fixed the verification/global_ocean.90x40x15 experiment: |
| 267 |
|
- new bathymetry (the world according to A., JMC, and M.) |
| 268 |
|
- new initial fields and forcing fields (*.bin files) |
| 269 |
|
- new POLY3.COEFFS (for the next release one should switch to a full |
| 270 |
|
equation of state: JMD95P or MDJWF) |
| 271 |
|
- fixed several errors and redundancies in the data file |
| 272 |
|
- experiment uses looped cells |
| 273 |
|
- added matlab directory with diagnostic scripts for plotting of output |
| 274 |
|
|
| 275 |
|
o S/R aim_initialise.F replace S/R aim_init from file aim_do_inphys.F: |
| 276 |
|
- read AIM physics parameters from a file (data.aimphys) |
| 277 |
|
- set defaults values = F.Molteni paper (Clim.Dyn., 2002) |
| 278 |
|
|
| 279 |
|
checkpoint46j_post |
| 280 |
|
o split calc_exact_eta in 2 S/R : integr_continuity & update_etaH |
| 281 |
|
o move wVel computation at the end of the time step, in S/R integr_continuity |
| 282 |
|
o create specific S/R to exchange T,S before DYNAMICS (for stagger time step) |
| 283 |
|
o update timeave pkg for wVel diagnostic. |
| 284 |
|
checkpoint46j_pre |
| 285 |
|
|
| 286 |
|
checkpoint46i_post |
| 287 |
|
o Clean up AIM package (and keep the results unchanged): |
| 288 |
|
include CPP_OPTION and use IMPLICT NONE in all routines ; |
| 289 |
|
declare all the variables _RL ; use _d 0 for all numerical constants. |
| 290 |
|
use ifdef ALLOW_AIM everywhere. And now AIM can be used with g77 ! |
| 291 |
|
|
| 292 |
checkpoint46h_post |
checkpoint46h_post |
| 293 |
o cleaned up the use of rhoNil and rhoConst. |
o cleaned up the use of rhoNil and rhoConst. |
Parent Directory
| 
Revision Log
| 
Revision Graph
| 
Patch