4 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
5 |
============================== |
============================== |
6 |
|
|
7 |
|
o pkg/exf (interpolation): |
8 |
|
- extended input field (2 rows) near the N & S poles: |
9 |
|
* fill in with the symetric value (when even Nb of data in longitude) |
10 |
|
* add average value at the poles only for scalar quantities ; for vector |
11 |
|
component interpolation, skip the averaging and keep duplicated values, |
12 |
|
unless the 2 components have same location (uvInterp=T) in which case |
13 |
|
interpolate the 2 components together using S/R EXF_INTERP_UV. |
14 |
|
- new S/R EXF_INTERP_UV to interpolate the 2 components of a vector field |
15 |
|
together: this allows to account properly for local orientation when |
16 |
|
averaging at N & S pole. |
17 |
|
- add CPP option EXF_USE_OLD_INTERP_POLE to recover old results regarding |
18 |
|
exf-interpolation near N & S poles. Undef by default. |
19 |
|
- add debug check/print on 1rst iter if debugLevel >=2. |
20 |
|
o pkg/exf: |
21 |
|
- change masking flags definition (no longer "parameter"), put them in |
22 |
|
common block and set them in S/R EXF_INIT_FIXED. |
23 |
|
- change recognized masking flag from s,u,v to c,w,s since the mask |
24 |
|
to use (maskC,W,S) depends on position on model grid and has little to |
25 |
|
do with scalar/vector difference; fix masking if using P-coordinates. |
26 |
|
- fix masking of wind-stress (was wrong if interpolated or read on A-grid) |
27 |
|
and update results of exp. global_with_exf (x2). |
28 |
|
- fix setting of zenith-angle table when SHORTWAVE_HEATING is undef. |
29 |
|
o pkg/obcs |
30 |
|
- S/R obcs_calc_stevens: fix a small bug (gammat -> gammas where appropriate) |
31 |
|
o pkg/exf (interpolation): |
32 |
|
- assume periodicity in X only if input field cover full longitude range. |
33 |
|
- improve search for lat. index (supposed to be faster, in ~log2(ny) steps, |
34 |
|
and should vectorise). |
35 |
|
- fix input lat of the 2 added row (in case we provide N.pole data). |
36 |
|
|
37 |
|
checkpoint63h (2011/12/30) |
38 |
|
o pkg/seaice: |
39 |
|
- change declaration of UVM from _RL to _RS in order to get B-grid |
40 |
|
uIce,vIce monitor when _RS = real*4 |
41 |
|
o pkg/offline: |
42 |
|
- new S/R OFFLINE_RESET_PARMS for resetting main-model params (switch off flags); |
43 |
|
- compute vertical velocity if not read from file (switch on exactConserv |
44 |
|
& staggerTimeStep); |
45 |
|
- fix output of ptracer monitor iter & time for default PTRACERS_monitorFreq. |
46 |
|
o verification: |
47 |
|
- update grid-files (in tutorial_held_suarez_cs/input and fizhi-cs-32x32x40/input) |
48 |
|
with grid-angle generated using "utils/matlab/cs_grid/cubeCalcAngle.m": |
49 |
|
fix Pb in yG definition @ 2 missing corners (previously yG was averaged); |
50 |
|
- update output of exp. global_ocean.cs32x15.viscA4 after changing |
51 |
|
grid-file angles (used here for coriolis-3d); |
52 |
|
- remove (or gzip) few files not needed to run fizhi-cs experiments. |
53 |
|
o pkg/exf (vector interpolation): |
54 |
|
- rotate vector component toward local grid-axis orientation using grid-angles |
55 |
|
(stored in GRID.h) instead of locally computed rotation angle; keep old code |
56 |
|
available with exf option EXF_USE_OLD_VEC_ROTATION (undef by default). |
57 |
|
o pkg/seaice: |
58 |
|
- allow to compile with exf options ALLOW_ATM_TEMP or ALLOW_DOWNWARD_RADIATION |
59 |
|
undefined. Fix and remove the stop when ALLOW_RUNOFF is undefined. |
60 |
|
o pkg/PKG_OPTIONS.h files: |
61 |
|
- move #include "CPP_OPTIONS.h" just after #include "PACKAGES_CONFIG.h", |
62 |
|
outside of the #ifdef ALLOW_THISPKG / #endif block. |
63 |
|
- update verification/*/code/PKG_OPTIONS.h files. |
64 |
|
o add a verification experiment (offline_exf_seaice/input.seaice) that test |
65 |
|
the B-grid LSOR solver |
66 |
|
o pkg/seaice: sort out emissivity and long wave radiation step 2: |
67 |
|
- add emissivity for snow and ice as new run-time parameters |
68 |
|
(unfortunately) that default to the corresponding exf values; this |
69 |
|
changes results in most verification experiments so that for backward |
70 |
|
compatibility these parameter have to be set explicitly to inconsistent |
71 |
|
values (ugly); |
72 |
|
- the exf-flag EXF_LWDOWN_WITH_EMISSIVITY then gets rid off the hard-wired |
73 |
|
emissivities of 0.97 associated with lwdownloc in S/R seaice_solve4temp |
74 |
|
until we agree how to handle this bug properly |
75 |
|
o pkg/exf (interpolation): |
76 |
|
- add a set of run-time param ({inputfield}_interpMethod), one for |
77 |
|
each interpolated input field, to select the interpolation method |
78 |
|
with a value of zero switching off the interpolation; |
79 |
|
- print interpolation parameters to standard output; |
80 |
|
- refine exf_check.F accordingly ; stop if USE_NO_INTERP_RUNOFF is defined |
81 |
|
(to use instead runoff_interpMethod=0 in data.exf). |
82 |
|
- simplify default setting of interpolation parameters (for all fields, |
83 |
|
set position to match model grid cell-center position, leading to trivial |
84 |
|
interpolation in case of regular Lat-Lon grid) and fix for the case of |
85 |
|
non-uniform delY. |
86 |
|
o pkg/exf/exf_radiation.F: add the factor ocean_emissivity to lwdown, for |
87 |
|
within #ifdef EXF_LWDOWN_WITH_EMISSIVITY/#endif. Unless ocean_emissivity=1. |
88 |
|
this factor is absolutely required, because the lw exitance (= out-going |
89 |
|
long wave radiation) is emissivity*stefanBoltzmann*T^4 + rho*lwdown, |
90 |
|
where the reflectivity rho = 1-emissivity for conservation reasons. Once |
91 |
|
this is established, we can remove the CPP-Flag (if we want to). |
92 |
|
o pkg/seaice: |
93 |
|
- qualify a stop statement in S/R seaice_check: EXF_READ_EVAP and |
94 |
|
ALLOW_BULKFORMULAE are only required if SEAICE_EXTERNAL_FLUXES is defined |
95 |
|
- sort out emissivity and long wave radiation, step 1: make SEAICE_emissivity |
96 |
|
what the name implies (and not emissivity*BoltzmannConstant), this means |
97 |
|
the old default value of 5.5e-8 now means |
98 |
|
5.5e-8/5.670e-8 = 0.97001763668430343479 as ocean_emissivity in |
99 |
|
S/R exf_readparms |
100 |
|
- FV-discretization of the B-grid LSOR solver, turn on with |
101 |
|
SEAICE_LSRBNEW. This new solver also involves a slight re-organization |
102 |
|
of the calling routine dynsolver, so that B and C-grid code look a |
103 |
|
little more alike. |
104 |
|
- fix index bugs in ostres.F (B-grid code) |
105 |
|
o pkg/seaice/seaice_ocean_budget.F: change (unused) code to make it look better |
106 |
|
o pkg/seaice/seaice_readparms.F: make ocean_emissivity*stefanBoltzmann the |
107 |
|
default for SEAICE_emissivity, if useEXF; requires setting a wrong |
108 |
|
parameter value in global_ocean.cs32x15/input.seaice/data.seaice for |
109 |
|
backward compatiblity |
110 |
|
o model/src: use sigmaR in calc_ivdc |
111 |
|
|
112 |
|
checkpoint63g (2011/12/16) |
113 |
|
o pkg/seaice/seaice_growth.F: |
114 |
|
fix heat conservation of the coupled ocean-seaice system. |
115 |
|
when SEAICE_HEAT_CONSERV_FIX is defined, the mismatch of |
116 |
|
solid vs liquid water treatment is resolved via QNET. |
117 |
|
o pkg/seaice/seaice_growth.F: |
118 |
|
add missing term for sublimation in saltFlux, which was breaking conservation. |
119 |
|
o pkg/seaice/seaice_model.F: |
120 |
|
Remove this patholigical case treatment in evolution branch, |
121 |
|
where it is un-necessary and breaks conservation. |
122 |
|
Update global_ocean.cs32x15/results accordingly for |
123 |
|
output.seaice.txt output_adm.seaice.txt and output_adm.seaice_dynmix.txt |
124 |
|
o model/src: |
125 |
|
useRealFreshWaterFlux: remove masking of EmPmP (using maskInC) in |
126 |
|
external_forcing_surf.F and apply mask directly in S/R SOLVE_FOR_PRESSURE. |
127 |
|
o pkg/obcs: |
128 |
|
modify Stevens boundary conditions so that S/R obcs_calc_stevens computes |
129 |
|
tendency for temperature and salinity that is then applied to update |
130 |
|
theta and salt in S/R obcs_apply_ts; this requires proper masking of all |
131 |
|
horizontal advection and diffusion contributions on the boundary (in |
132 |
|
S/Rs gad_advection, gad_calc_rhs, gmredi_rtransport, gmredi_calc_diff), and |
133 |
|
removing the mask in S/R timestep_tracer. The advantage of this approach is |
134 |
|
that the updated tracer values include contributions from the surface |
135 |
|
forcing and vertical diffusion. |
136 |
|
This modification changes exp4.stevens because of vertical diffusion. |
137 |
|
o model/src: |
138 |
|
change non-curvilinear grid setting to also work with exch2 + blank tiles: |
139 |
|
- move delX,delY to new header file (SET_GRID.h) and adjust length to 1rst |
140 |
|
face dimensions. |
141 |
|
- new S/R INI_LOCAL_GRID to define model tile-local grid (using same units |
142 |
|
as delX,delY) and called from S/R INI_CARTESIAN_GRID, INI_CYLINDER_GRID |
143 |
|
and INI_SPHERICAL_POLAR_GRID. |
144 |
|
change test-experiment global_ocean.90x40x15: |
145 |
|
- split the domain in many more tiles (9x4 tiles of 10x10 each); |
146 |
|
- switch to exch2 and remove 1 blank-tile (but only for MPI test). |
147 |
|
- update results of both test (std & dwnslp). |
148 |
|
o pkg/diagnostics/diagstat_calc.F: fix problem with non-initialised halos in |
149 |
|
tmp-fields for TARGET_NEC_SX at the cost of extra loops and less vectorisation |
150 |
|
o model/src: |
151 |
|
- take sigma-coords related code out of update_etah.F into new S/R UPDATE_ETAWS |
152 |
|
and fix initialisation when using sigma-coords. |
153 |
|
o model/src: |
154 |
|
- improve Gael's changes from Jan 14, 2011 "addition of an 'update' of hfacc etc. |
155 |
|
to their current value at the beginning of forward_step.F": add a new run-time |
156 |
|
switch "doResetHFactors" (default=false) and use it to do (or not) the resetting. |
157 |
|
- set doResetHFactors=T in the following test exp: |
158 |
|
global_ocean.90x40x15/input (rStar) & input.dwnslp (linFS), |
159 |
|
global_ocean.cs32x15/input.seaice (rStar), hs94.cs-32x32x5/input.impIGW (linFS) |
160 |
|
and tutorial_global_oce_in_p (nonLinFS in p) |
161 |
|
o pkg/ecco |
162 |
|
- added IES cost term |
163 |
|
o verification/advect_xz |
164 |
|
- scale down the depth by 1/100 (from 200 km to just 2.km) |
165 |
|
- add secondary test using divergent flow, r-star and implicit vertical advection. |
166 |
|
- add note (check_conserve_TS.txt) and matlab script to check global |
167 |
|
conservation of T & S with NonLin-FreeSurf and Adams-Bashforth. |
168 |
|
o pkg/rbcs: |
169 |
|
- add missing relaxUFile & relaxVFile in namelist (in rbcs_readparms.F). |
170 |
|
o pkg/obcs (Stevens BCs): |
171 |
|
- fix various problems in the phase speed computations of the Stevens |
172 |
|
boundary conditions: |
173 |
|
- sign error for western and southern boundaries, |
174 |
|
- time tendencies were underestimated by factor of dTracerLev and some just |
175 |
|
zero (for multidim advection schemes) -> now estimate time tendencies |
176 |
|
from difference between previous (n) time level and the one before (n-1); |
177 |
|
this requires storing a field per tracer per boundary, also in the pickup |
178 |
|
files. |
179 |
|
- replace deltaTmom in cflMer/Zon computation with dTracerLev |
180 |
|
- set default phase velocity to cflMer/Zon |
181 |
|
- update verification experiment exp4.stevens |
182 |
|
o pkg/generic_advdiff (implicit_r): |
183 |
|
- fix implicit vertical advection conservation with AB and/or NonLin-FreeSurf |
184 |
|
(remove Tr*d/dz(w) in gad_implicit_r.F and add it in gad_calc_rhs.F); |
185 |
|
- fix implicit vertical diffusion or advection with NonLin FreeSurf |
186 |
|
(was using current recip_hFacC instead of future recip_hFacC): set future |
187 |
|
recip_hFac in thermodynamics.F and pass & use it in all vertical implicit |
188 |
|
solver S/R ; |
189 |
|
- fill diagnostics of vertical advective fluxes (computed from tendency) |
190 |
|
when using implicit vert. advection; |
191 |
|
- update output of exp that uses NonLin-FreeSurf & implicit vertical diffusion: |
192 |
|
global_ocean.90x40x15 (fwd+AD), global_ocean.cs32x15 (4 fwd +4 AD), |
193 |
|
tutorial_global_oce_in_p (fwd), tutorial_tracer_adjsens (AD) |
194 |
|
and cpl_aim+ocn (used in ocean); to be fixed: bottom_ctrl_5x5. |
195 |
|
o pkg/diagnostics: |
196 |
|
- add diags for T & S tendency which goes through Adams-Bashforth. |
197 |
|
|
198 |
|
checkpoint63f (2011/11/30) |
199 |
|
o pkg/obcs/obcs_calc_stevens.F: |
200 |
|
- join some loops to reduce number of if-statements |
201 |
|
- fix a bug in code for northern and southern boundaries. |
202 |
|
o pkg/seaice/seaice_init_varia.F |
203 |
|
- undo resetting UVM=0 that has slipped in unnoticed in r1.49 and |
204 |
|
caused zero ice-velocities for the B-grid code |
205 |
|
o pkg/seaice/seaice_init_fixed.F |
206 |
|
- fix a bug in the initialisation of the second metric terms |
207 |
|
k2AtC and k2AtZ: k2AtC was set to the value of k2AtZ and k2AtZ was 0. |
208 |
|
- this affects all seaice configurations with useCurviLinearGrid = .true. |
209 |
|
and SEAICEuseDynamics = .true. (default) and SEAICEuseMetricTerms |
210 |
|
= .true. (default) |
211 |
|
- the only verification experiment affected is global_ocean.cs32x15: |
212 |
|
icedyn, seaice, and seaice_dynmix (AD) |
213 |
|
o pkg/thsice/thsice_diagnostics_state.F |
214 |
|
- only for TARGET_NEC_SX: initialise all of tmpFld, so that overlap does |
215 |
|
not contain garbage that can lead to surprises in diagstats_calc.F |
216 |
|
o model/src: |
217 |
|
- move bi,bj loops inside INTEGR_CONTINUITY ; |
218 |
|
- call INTEGR_CONTINUITY directly from forward_step.F |
219 |
|
(previously called from momentum_correction_step.F) |
220 |
|
- call UPDATE_ETAH from integr_continuity.F and update initialise_varia.F, |
221 |
|
forward_step.F and pkg/ecco/the_main_loop.F |
222 |
|
o model/src/do_oceanic_phys.F: |
223 |
|
- reset addMass to zero only if useICEFRONT ; |
224 |
|
- separated k-loop for rhoInSitu calculation ; |
225 |
|
- add 2 diagnostics for wVel*delta^k(Rho) at constant pressure (= WdRHO_P) |
226 |
|
and at constant T,S (= WdRHOdP); |
227 |
|
NOTE: change diagnostics WRHOMASS : this is now simply the Vertical transport |
228 |
|
of In-Situ Density (Anomaly) (=wVel*rho^bar-k) (previously was Vertical |
229 |
|
transport of locally-referenced potential density); |
230 |
|
to recover old diag: (WRHOMASS)_old = (WRHOMASS)_new -1/2 * (WdRHOdP) |
231 |
|
o gluemnc: |
232 |
|
- replace ncrcat with ncks when adding variables to *.glob.nc |
233 |
|
- change the way variables without record dimensions are handled so that |
234 |
|
files that have variables with and without record variables can be glued: |
235 |
|
add a record dimension with a new unique name to be able to differentiate |
236 |
|
between variable with and without a-prior record dimensions. Remove |
237 |
|
this record dimension before adding the variable to the *.glob.nc file. |
238 |
|
This procedure automatically works also for grid.*.nc files that do not |
239 |
|
have any record dimension. |
240 |
|
o verification/cpl_aim+ocn: |
241 |
|
- generate new reference output on baudelaire (gfortran, mpich2) (previously |
242 |
|
from old aces-grid cluster, ao + ifort) and update 'run_cpl_test' script. |
243 |
|
o pkg/obcs/obcs_calc_stevens.F: |
244 |
|
- add code for northern and southern boundaries. |
245 |
|
- reorder some loops to save some operations and so that drFBar can be scalar |
246 |
|
o model/src: |
247 |
|
- re-arrange extrapolation calculation in Adams-Bashforth S/R (affects machine |
248 |
|
truncation) and add diagnostics for Adams-Bashforth tendency increment. |
249 |
|
- update test-exp. AD output: |
250 |
|
1D_ocean_ice_column, bottom_ctrl_5x5, lab_sea.noseaicedyn, obcs_ctrl; |
251 |
|
and FWD output: exp4 (x3), fizhi-cs-32x32x40, fizhi-cs-aqualev20, |
252 |
|
front_relax (x3), global_ocean.90x40x15 (x2), global_ocean.cs32x15 (x5), |
253 |
|
global_ocean_ebm, ideal_2D_oce, internal_wave, isomip (x3), lab_sea (x4), |
254 |
|
matrix_example, MLAdjust (x5), natl_box (x2), tutorial_advection_in_gyre, |
255 |
|
tutorial_global_oce_in_p, tutorial_global_oce_latlon and |
256 |
|
tutorial_plume_on_slope. |
257 |
|
o pkg/diagnostics/diagstats_calc.F: add special code for TARGET_NEC_SX |
258 |
|
that vectorises. This code uses 3 extra 2D fields, F90 array assignments |
259 |
|
and F90 intrinsic routines SUM, MINVAL, MAXVAL, but since TARGET_NEC_SX |
260 |
|
does not even come with an F77 compiler anymore, this should not be |
261 |
|
a limitation |
262 |
|
|
263 |
|
checkpoint63e (2011/11/06) |
264 |
|
o pkg/seaice adjoint of LSR: |
265 |
|
- substantially rework STORE dirs. |
266 |
|
- add new flag SEAICE_LSR_ADJOINT_ITER and parameter SOLV_MAX_FIXED |
267 |
|
this flag will do SOLV_MAX_ITERS but only store SOLV_MAX_FIXED fields |
268 |
|
and only perform SOLV_MAX_FIXED adjoint iters. |
269 |
|
If flag SEAICE_LSR_ADJOINT_ITER is undef-ined, the store in the iteration |
270 |
|
always goes to same record, meaning that the last forward field will be |
271 |
|
the one used in the adjoint - an approximation. |
272 |
|
- now produces non-zero LSR-related gradients, and acceptable gradients for |
273 |
|
An's zonal (x-z) test case (to be checked in) |
274 |
|
- lab_sea: produces acceptable gradients for lab_sea with forward/backward LSR |
275 |
|
(but these are screwed up by the thermodynamics in the first place); |
276 |
|
reduce from 4 to 3 timesteps to avoid t.d. issue as temporary solution; |
277 |
|
output_adm.txt also reflects changes to seaice_lsr |
278 |
|
- global_ocean.cs32x15: |
279 |
|
* the "default" input_ad.seaice has been re-defined to always exclude dynamics |
280 |
|
(output_adm.seaice.txt was updated) |
281 |
|
* new input_ad.seaice_dynmix tests a forward LSR / backward free-drift |
282 |
|
(output_adm.seaice_dynmix.txt) |
283 |
|
o test-exp. exp4+stevens: |
284 |
|
- switch off pTracers and nonHydrostatic (restart problems with NH); update output |
285 |
|
o pkg/seaice (lsr solver): |
286 |
|
- put coriolis term into FORCEX,Y (as in seaice_evp.F; previously in rhsU,V only); |
287 |
|
update output (change truncation) of FWD test exp lab_sea & global_ocean.cs32x15 |
288 |
|
- fix SEAICEuseFREEDRIFT for new OBCS version (ndef OBCS_UVICE_OLD) |
289 |
|
- new parameter "LSR_mixIniGuess" allows (if=2,4) to mix free-drift solution |
290 |
|
into initial guess for linearised solver; use weight ~ (1/err)^2,4. |
291 |
|
- compute (with new S/R) and print initial (+ free-driff sol. if available) |
292 |
|
and final residual of linearised system. |
293 |
|
o pkg/pp81: add diagnostics for vertical eddy viscosity and diffusivity |
294 |
|
(requires new subroutines pp81_init_fixed.F/pp81_diagnostics_init.F) |
295 |
o pkg/obcs: Stevens BCs (still only for east and west) |
o pkg/obcs: Stevens BCs (still only for east and west) |
296 |
- fix Stevens BCs by using intermediate velocities of previous |
- fix Stevens BCs by using intermediate velocities of previous |
297 |
time step as "baroclinic" contribution rather velocity of one grid cell |
time step as "baroclinic" contribution rather velocity of one grid cell |
298 |
inside the boundaries; more consistent but makes the entire treatment |
inside the boundaries; more consistent but makes the entire treatment |
299 |
more complicated, requires new pickup files and two new S/Rs: |
more complicated, requires new pickup files and two new S/Rs: |
300 |
obcs_copy_uv_n and obcs_save_uv_n |
obcs_copy_uv_n and obcs_save_uv_n |
301 |
keep old version with flag OBCS_STEVENS_USE_INTERIOR_VELOCITY |
keep old version with flag OBCS_STEVENS_USE_INTERIOR_VELOCITY |
302 |
(undefined by default) |
(undefined by default) |
303 |
- silently move computation of some grid factors into separate loops to |
- silently move computation of some grid factors into separate loops to |
304 |
save some recomputations |
save some recomputations |
305 |
- add a hack for non-linear free surface |
- add a hack for non-linear free surface |
306 |
- add a verification experiment to exp4 that tests this code |
- add a verification experiment to exp4 that tests this code |