| 3 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
| 4 |
============================== |
============================== |
| 5 |
|
|
| 6 |
|
checkpoint50a_post |
| 7 |
|
o Bug fix for merging between c50 and KPP. |
| 8 |
|
ikey was passed from thermodynamics to kpp_calc via |
| 9 |
|
common block rather than being recomputed in kpp_calc, |
| 10 |
|
in contradiction with new key itdkey. |
| 11 |
|
New key ikppkey created, and tamc.h headers updated. |
| 12 |
|
|
| 13 |
|
checkpoint50 |
| 14 |
|
- 2nd step of merge with ecco-branch: |
| 15 |
|
merge changes in ecco-branch back into main trunk |
| 16 |
|
- bug fix in find_alpha |
| 17 |
|
- more storing in gmredi_calc_tensor.F, kpp_routines.F |
| 18 |
|
- avoid iterative loop in ini_pressure for ALLOW_AUTODIFF_TAMC |
| 19 |
|
- merged some sliced AD support routines |
| 20 |
|
|
| 21 |
|
checkpoint49 |
| 22 |
|
merged from ecco-branch: |
| 23 |
|
o exf: |
| 24 |
|
- Enable initialisation of forcing fields to constant |
| 25 |
|
(runtime) values. |
| 26 |
|
- in exf_getffields.F |
| 27 |
|
Reduce i-/j-loop to interior domain, discarding overlaps. |
| 28 |
|
That also fixes wrong TAF-key computations for key_1, key_2 |
| 29 |
|
with bulf formulae. |
| 30 |
|
- exf_init.F modify #ifdef for exf_init_evap |
| 31 |
|
- exf_getffieldrec.F, ctrl_getrec.F |
| 32 |
|
The following INT-usages are not safe: |
| 33 |
|
fldsecs = int(fldsecs/fldperiod)*fldperiod |
| 34 |
|
fldcount = int(fldsecs/fldperiod) + 1 |
| 35 |
|
and were modified. |
| 36 |
|
o autodiff: checkpoint_lev... |
| 37 |
|
- updated storing for bulk formulae (some arrays were missing) |
| 38 |
|
- added state for seaice package |
| 39 |
|
- synchronized with main branch |
| 40 |
|
o DIVA: |
| 41 |
|
Modif.s in |
| 42 |
|
- the_main_loop.F |
| 43 |
|
- the_model_main.F |
| 44 |
|
and added cost_final_restore.F cost_final_store.F |
| 45 |
|
o With these changes, it is possible to fully separate |
| 46 |
|
the first forward run to compute the cost function (mdthe_main_loop) |
| 47 |
|
and adjoint runs over individual DIVA intervals (adthe_main_loop) |
| 48 |
|
into separate model executions. |
| 49 |
|
This enables better tuning of DIVA to given queue sizes. |
| 50 |
|
o Loss of cost function value in consecutive DIVA runs is |
| 51 |
|
fixed by storing to 'costfinal' file. |
| 52 |
|
o Appropriate call of ctrl_unpack/_pack,grdchk_main at |
| 53 |
|
beginning/end of adjoint integration is accounted for. |
| 54 |
|
The current changes apply to TAF-1.4.26. |
| 55 |
|
TAF has one remaining bug, requiring one manual change |
| 56 |
|
(see MITgcm manual). |
| 57 |
|
o cal: synchronized between MAIN and ecco-branch |
| 58 |
|
|
| 59 |
|
checkpoint48i_post |
| 60 |
|
o Changes to restore differentiability of code w.r.t. previous tag |
| 61 |
|
(mostly adding new routines to make list and replacing |
| 62 |
|
pressure by totPhiHyd). |
| 63 |
|
o Updated and merged grdchk package |
| 64 |
|
- has both ADM and TLM checks |
| 65 |
|
- works for single- and multi-proc. |
| 66 |
|
- output cleaned |
| 67 |
|
- worked successfully for parallel DIVA |
| 68 |
|
o genmake: |
| 69 |
|
- Adding -platform=SP4 |
| 70 |
|
- Modif. for DEC Alpha (many don't have makedepend; use mkdep instead). |
| 71 |
|
o Modif. for DIVA: |
| 72 |
|
ctrl_unpack should not be called in intermediate DIVA interval. |
| 73 |
|
(similar thing would apply for active files such as |
| 74 |
|
tbar, sbar, psbar, which should not be reinitialised; |
| 75 |
|
only relevant for ecco-branch) |
| 76 |
|
|
| 77 |
|
checkpoint48h_post |
| 78 |
|
o improve "ini_p_ground" when using finite difference form for calc_phi_hyd |
| 79 |
|
o Added net flux to downward flux conversion to pkg/exf/exf_getffields.F |
| 80 |
|
o Added SEAICE_initialHEFF to pkg/seaice |
| 81 |
|
o ecco_the_model_main.F: SEAICE_WRITE_PICKUP must precede WRITE_CHECKPOINT |
| 82 |
|
|
| 83 |
|
checkpoint48g_post |
| 84 |
|
--- EOS: |
| 85 |
|
o compute locally the pressure for use in EOS: UNESCO, JMD95P, MDJWF |
| 86 |
|
store total Potential in totPhyHyd for diagnostics & EOS. |
| 87 |
|
o fix restart and overlap Pb when using Z-coord and EOS funct. of P |
| 88 |
|
--- Pressure/geopotential gradient: |
| 89 |
|
o use Ro_surf & R_low instead of hFac in cal_phi_hyd: |
| 90 |
|
o NLFS + p-coord.: eta*Alpha' is not longer lagging 1.time-step behind |
| 91 |
|
( changes results of global_ocean_pressure ) |
| 92 |
|
o NLFS + z-coord.: add eta*Rho' contribution. |
| 93 |
|
o accurate phiHyd when using interface-W at the middle between 2 centers |
| 94 |
|
( changes results of ideal_2D_oce ) |
| 95 |
|
o includes r* 2nd term in potential gradient. |
| 96 |
|
|
| 97 |
|
checkpoint48f_post |
| 98 |
|
Merging from release1_p12: |
| 99 |
|
o Modifications for using pkg/exf with pkg/seaice |
| 100 |
|
- improved description of the various forcing configurations |
| 101 |
|
- added basic radiation bulk formulae to pkg/exf |
| 102 |
|
- units/sign fix for evap computation in exf_getffields.F |
| 103 |
|
- updated verification/global_with_exf/results/output.txt |
| 104 |
|
o Added pkg/sbo for computing IERS Special Bureau for the Oceans |
| 105 |
|
(SBO) core products, including oceanic mass, center-of-mass, |
| 106 |
|
angular, and bottom pressure (see pkg/sbo/README.sbo). |
| 107 |
|
o Lower bound for viscosity/diffusivity in pkg/kpp/kpp_routines.F |
| 108 |
|
to avoid negative values in shallow regions. |
| 109 |
|
- updated verification/natl_box/results/output.txt |
| 110 |
|
- updated verification/lab_sea/results/output.txt |
| 111 |
|
o MPI gather, scatter: eesupp/src/gather_2d.F and scatter_2d.F |
| 112 |
|
o Added useSingleCpuIO option (see PARAMS.h). |
| 113 |
|
o Updated useSingleCpuIO option in mdsio_writefield.F to |
| 114 |
|
work with multi-field files, e.g., for single-file pickup. |
| 115 |
|
o pkg/seaice: |
| 116 |
|
- bug fix in growth.F: QNET for no shortwave case |
| 117 |
|
- added HeffFile for specifying initial sea-ice thickness |
| 118 |
|
- changed SEAICE_EXTERNAL_FLUXES wind stress implementation |
| 119 |
|
o Added missing /* */ to CPP comments in pkg/seaice, pkg/exf, |
| 120 |
|
kpp_transport_t.F, forward_step.F, and the_main_loop.F |
| 121 |
|
o pkg/seaice: |
| 122 |
|
- adjoint-friendly modifications |
| 123 |
|
- added a SEAICE_WRITE_PICKUP at end of the_model_main.F |
| 124 |
|
|
| 125 |
|
checkpoint48e_post |
| 126 |
|
o New S/R for diagnostic of bottom pressure (phi0surf was missing in c48d_post) |
| 127 |
|
o dynamics: change definition of computational domain & adapt mom_fluxform |
| 128 |
|
accordingly ; when viscA4=0, allows to run the dynamics with Olx=Oly=2. |
| 129 |
|
o following exp works with Olx=Oly=2 (and pass through checkbounds option): |
| 130 |
|
adjust_nlfs.cs-32x32x1 adjustment.128x64x1 adjustment.cs-32x32x1 |
| 131 |
|
exp0 exp1 exp2 exp4 exp5 hs94.cs-32x32x5 hs94.128x64x5 hs94.1x64x5 |
| 132 |
|
global_ocean.90x40x15 global_with_exf global_ocean_pressure |
| 133 |
|
inverted_barometer solid-body.cs-32x32x1 |
| 134 |
|
(high order AdvScheme and/or multiDimAd works with Olx=Oly=2 ? not sure) |
| 135 |
|
|
| 136 |
|
checkpoint48d_post |
| 137 |
|
o in preparation for r*: |
| 138 |
|
a) use pre-computed gradient of hydrostatic potential: |
| 139 |
|
changes in timestep.F & mom_cdscheme.F affects results of ideal_2D_oce |
| 140 |
|
b) move phi0surf from calc_phi_hyd to calc_grad_phi_hyd: |
| 141 |
|
=> affects results of glob_oce_pressure (different truncation error) |
| 142 |
|
update output.txt of ideal_2D_oce & global_ocean_pressure |
| 143 |
|
|
| 144 |
|
checkpoint48d_pre |
| 145 |
|
o in preparation for r*: |
| 146 |
|
new S/R (calc_grad_phi_hyd.F) to compute Hydrostatic potential gradient. |
| 147 |
|
pass the 2 comp. of the grad. as arguments to momentum S/R. |
| 148 |
|
but for the moment, only used if it does not change the results. |
| 149 |
|
o add finite volume form of calc_phi_hyd for buoyancyRelation OCEANIC & OCEANICP |
| 150 |
|
|
| 151 |
|
checkpoint48c_post |
| 152 |
|
Modifications in treatment of initial flux forcing controls |
| 153 |
|
to avoid extra storing of flux fields. |
| 154 |
|
|
| 155 |
|
checkpoint48c_pre |
| 156 |
|
o Added tidal_basin_2d experiment (currently 1D!) |
| 157 |
|
o can use r* Coordinate with OBC (but Orlanski Rad. BC not yet implemented) |
| 158 |
|
|
| 159 |
|
checkpoint48b_post |
| 160 |
|
o r* coordinate added in #ifdef NONLIN_FRSURF block. |
| 161 |
|
(modification to pressure gradient not yet implemented) |
| 162 |
|
o mom_fluxform: new S/R mom_calc_rtrans.F computes vertical transports |
| 163 |
|
(for U & V) that are used for vertical advection of momentum. |
| 164 |
|
|
| 165 |
|
checkpoint48a_post |
| 166 |
|
o In preparation for r*, use temp. array to compute vertical transport |
| 167 |
|
in flux-form momentum pkg. => change results of few test-exp. |
| 168 |
|
Update output of: exp4, exp5, front_relax, global_ocean_pressure, |
| 169 |
|
lab_sea, natl_box & plume_on_slope |
| 170 |
|
|
| 171 |
|
checkpoint48 |
| 172 |
|
o fix small bug that was causing lot of warning (TIMER_STOP 'UPDATE_CG2D') |
| 173 |
|
|
| 174 |
|
checkpoint47j_post |
| 175 |
|
o add new exp. (ideal_2D_oce) to test "exotic" parameters and options. |
| 176 |
|
o moved adexch from forward_step to addummy_in_stepping |
| 177 |
|
o added exch in ctrl_map_... |
| 178 |
|
o aveFreq.=0. at end of cost_final to avoid overwrite in |
| 179 |
|
reverse checkpointing loop |
| 180 |
|
o Yet more changes in gmredi: |
| 181 |
|
* adgmredi_calc_tensor |
| 182 |
|
avoiding all recomputation of gmredi_slope_limit |
| 183 |
|
* adgmredi_x/y/rtransport |
| 184 |
|
added flag for excessive storing to avoid recomp. of |
| 185 |
|
u/v/rtans, dTdx/y/z |
| 186 |
|
-> this is not really necessary and very memory-consuming |
| 187 |
|
* adgmredi_slope_psi: |
| 188 |
|
consistency with gmredi_slope_limit in treatment of GM_slopeSqCutoff |
| 189 |
|
* gmredi_slope_limit |
| 190 |
|
re-activated full calculation of taperfct for case 'ac02' |
| 191 |
|
o updating verification/ adjoints of carbon, global_ocean.90x40x15 |
| 192 |
|
|
| 193 |
|
checkpoint47i_post |
| 194 |
|
o GM_EXCLUDE_() replace GM_TAPER_(): by default, all the code is compiled. |
| 195 |
|
o add GM_Small_Number & GM_slopeSqCutoff as run-time parameters (data.gmredi) |
| 196 |
|
default values = 1.D-12 & 1.D+48 (=> give same results as checkpoint47f_post) |
| 197 |
|
o cg2dTargetResWunit: define tolerance using deltaTmom (and not deltaTfreesurf) |
| 198 |
|
o gmredi: fix few bugs. |
| 199 |
|
|
| 200 |
|
checkpoint47h_post (duplicate tag for PH's c48 tag, in order to move c48) |
| 201 |
|
This will be a good checkpoint to plug into the ecco-branch. |
| 202 |
|
o few modif.'s to get latest version adjointed |
| 203 |
|
(mainly kick out code in ini_linear_phisurf) |
| 204 |
|
o modif's to run adjoint with exactConserv |
| 205 |
|
o case GM_BOLUS_ADVEC should be cleaned |
| 206 |
|
S/R gmredi_slope_psi should be cleaned |
| 207 |
|
o verification/carbon now has exactConserv=.TRUE. |
| 208 |
|
|
| 209 |
|
checkpoint47g_post |
| 210 |
|
o time-average AIM output: becomes independent from statvars time-average |
| 211 |
|
(can be used with #undef ALLOW_TIMEAVE) ; add new variables (e.g., RH) |
| 212 |
|
o After TAF bug fix: |
| 213 |
|
Tested adjoint of gmredi package. |
| 214 |
|
Tests were done with checkpoint47 rather than with latest |
| 215 |
|
checkpoint47f_post since the latter is broken for the adjoint |
| 216 |
|
verification/carbon/ contains both ADM and TLM gradient checks. |
| 217 |
|
Here they are, before they get changed/lost/stolen. |
| 218 |
|
Mostly modif.'s to fix numerical sensitivities. |
| 219 |
|
Gradient checks OK for |
| 220 |
|
- GM_taper_scheme: |
| 221 |
|
* clipping |
| 222 |
|
* ac02 |
| 223 |
|
* linear |
| 224 |
|
* glw91 |
| 225 |
|
* dm95 |
| 226 |
|
* ldd97 |
| 227 |
|
- GMREDI_OPTIONS: |
| 228 |
|
* GM_VISBECK_VARIABLE_K |
| 229 |
|
* GM_NON_UNITY_DIAGONAL |
| 230 |
|
* GM_EXTRA_DIAGONAL |
| 231 |
|
* GM_BOLUS_ADVEC |
| 232 |
|
in conjunction with data.gmredi parameters to be checked in |
| 233 |
|
in a few minutes under verification/carbon/code/ |
| 234 |
|
|
| 235 |
checkpoint47f_post |
checkpoint47f_post |
| 236 |
Merging from release1_p10: |
Merging from release1_p10: |
| 237 |
o modifications for using pkg/exf with pkg/seaice |
o modifications for using pkg/exf with pkg/seaice |
Parent Directory
| 
Revision Log
| 
Revision Graph
| 
Patch