4 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
5 |
============================== |
============================== |
6 |
|
|
7 |
|
o vermix test experiment: switch from LINEAR EOS to MDJWF (not tested before) |
8 |
|
o new pkg: down_slope |
9 |
|
density-driven downsloping flow parameterisation (Tellus, 1999, 412-430). |
10 |
|
- presently, modify T & S inland (dwnslp_calc_rho.F). |
11 |
|
can be avoided if rho was stored in common bloc and used in calc_phi_hyd. |
12 |
|
- implementation for ocean in P-coords not tested. |
13 |
|
- add secondary test to global_ocean.90x40x15 (input.dwnslp) to check this pkg. |
14 |
|
o move kSurfC,kSurfW & kSurfS from SURFACE.h to GRID.h (next to kLowC) |
15 |
|
(to facilitate conversion of oceanic pkg to p-coordinate). |
16 |
|
o change FIND_RHO to FIND_RHO_2D : |
17 |
|
pass arguments tFld & sFld as local (tile) 2D-arrays ; +fix EOS_CHECK. |
18 |
|
o new pkg: cheapaml |
19 |
|
|
20 |
|
checkpoint61b |
21 |
|
o exch2: |
22 |
|
change index bounds in rx2_cube exchanges (new S/R: EXCH2_GET_UV_BOUNDS) |
23 |
|
- no longer depend on the order sequence (N,S,E,W). |
24 |
|
- 3rd exchange no longer needed (tested with 24 tiles). |
25 |
|
- same modif to hand-written adjoint S/R (global_ocean.cs32x15: zero diff) |
26 |
|
- exch_UV_A-grid readily available (but not yet tested). |
27 |
|
comment out 3rd exch call and remove copy of V to U at SE & NE face-corners |
28 |
|
(both are no longer needed) in exch_uv: no effect in forward run; but |
29 |
|
get truncation error differences with adjoint exch. |
30 |
|
o exch2: |
31 |
|
- change index-bounds storage (moved from target to local tile, |
32 |
|
more intuitive this way) ; rename/remove some variables. |
33 |
|
- change utils/exch2/matlab-topology-generator & code_mods accordingly. |
34 |
|
|
35 |
|
checkpoint61a |
36 |
|
o exch2/matlab-topology-generator: |
37 |
|
fix bug in index bounds for N-S internal exchange (hard to tell if this |
38 |
|
had any effect on the solution). |
39 |
|
o pkg/mnc: small fix in mnc_handle_err: make sure that length of msgbuf |
40 |
|
is exceeded |
41 |
|
o testreport: modify syntax of awk, so that it works also with |
42 |
|
awk, and not only with nawk and gawk |
43 |
|
|
44 |
|
checkpoint61 |
45 |
|
o autodiff |
46 |
|
Change default for TAF usage by removing argument '-nonew_arg' |
47 |
|
This changes TAF subroutine argument list (ordering of active var.) |
48 |
|
Omission of this flag leads to incompatibility w.r.t. TAMC |
49 |
|
To restore TAMC-compatibility, need following modifs: |
50 |
|
* use adoptfile tools/adjoint_options/adjoint_tamc_compatibility |
51 |
|
* use CPP option #define AUTODIFF_TAMC_COMPATIBILITY |
52 |
|
Tested TAF version is 1.9.22 |
53 |
|
N.B.: exch2 hand-written adjoint code currently not TAMC compatible |
54 |
|
|
55 |
|
checkpoint60 |
56 |
|
--->>> This is the last checkpoint for which generating the adjoint |
57 |
|
--->>> is done via the TAMC-compatibility TAF-flag "-nonewarg" |
58 |
|
--->>> by default. |
59 |
|
--->>> Future checkpoints will ommit this flag by default. |
60 |
|
o pkg/generic_advdiff, os7mp: simplify and avoid division by zero. |
61 |
|
o pkg/grdchk: Add gradient check to etan0 perturbations. |
62 |
|
o pkg/exf: read ice area from file (#ifdef ALLOW_ICE_AREAMASK). |
63 |
|
o pkg/autodiff: more flexibility in zeroadj. |
64 |
|
model/src: modifying do_oceanic_phys.F accordingly. |
65 |
|
o pkg/ctrl: Bounds for (kapgm,kapredi,eddypsi) parameter estimation. |
66 |
|
o model/src and pkg/gmredi: GMREDI_WITH_STABLE_ADJOINT CPP option |
67 |
|
special setup of gmredi for which the adjoint is stable |
68 |
|
enough for parameter optimization. |
69 |
|
o bridging the gap between eddy stress and GM. |
70 |
|
-> eddyTau is replaced with eddyPsi (eddyTau = f x rho0 x eddyPsi) |
71 |
|
along with a change in CPP option (now ALLOW_EDDYPSI). |
72 |
|
-> when using GM w/ GM_AdvForm: |
73 |
|
The total eddy streamfunction (Psi = eddyPsi + K x Slope) |
74 |
|
is applied either in the tracer Eq. or in momentum Eq. |
75 |
|
depending on data.gmredi (intro. GM_InMomAsStress). |
76 |
|
-> ALLOW_EDDYPSI_CONTROL for estimation purpose. |
77 |
|
The key modifications are in model/src/taueddy_external_forcing.F |
78 |
|
pkg/gmredi/gmredi_calc_*F pkg/gmredi/gmredi_*transport.F |
79 |
|
|
80 |
|
checkpoint59r |
81 |
|
o new S/R (pkg/mdsio/mdsio_facef_read.F) reading of sNx+1 x sNy+1 array |
82 |
|
from per-face file (take S/R READSYMTILE_RS out of ini_curvilinear_grid.F); |
83 |
|
fix option to read coriolis(@ cell-corner) from files with CS-grid. |
84 |
|
o pkg/mnc: add some attributes to standard 1D-coordinate variables |
85 |
|
o pkg/diagnostics: enforce missing_value attribute for scalar variables at |
86 |
|
mass points, when using mnc for output |
87 |
|
- add runtime parameter (missing_value, missing_value_int) to override |
88 |
|
default (which is the value of getcon('UNDEF')) |
89 |
|
o pkg/mnc: prepare mnc for the actual use of missing values with the |
90 |
|
diagnostics package: small fix in mnc_cw_readwrite.template and more |
91 |
|
initialisation of never used fields |
92 |
|
- replace mnc_common.h and mnc_id_header.h with corresponding upper case |
93 |
|
versions |
94 |
|
o pkg/thsice/thsice_ini_vars.F added default initial values so that |
95 |
|
thermodynamics sea ice solver does not blow up when some initial fields |
96 |
|
are missing. |
97 |
|
o move from external_forcing_surf.F to ini_forcing.F |
98 |
|
the p-coord. hack consisting in using pLoad to set phi0surf ; |
99 |
|
update output of tutorial_global_oce_in_p (affected @ truncation level). |
100 |
|
o Freshen up a bit hs94.1x64x5, notably: |
101 |
|
remove rotationPeriod and tRef=5*0. (adjoint only), and add |
102 |
|
exactConserv=.TRUE. |
103 |
|
o pkg/generic_advdiff: |
104 |
|
change option GAD_ALLOW_SOM_ADVECT to GAD_ALLOW_TS_SOM_ADV which only |
105 |
|
applies to files where Temperature & Salinity 2nd Order moments are used. |
106 |
|
o pkg/ptracers: add second-order moment advection schemes (80 and 81); |
107 |
|
this uses a dynamically allocated internal state data structure |
108 |
|
(#define PTRACERS_ALLOW_DYN_STATE in PTRACERS_OPTIONS.h) |
109 |
|
and requires a fortran 90 compiler |
110 |
|
|
111 |
|
checkpoint59q |
112 |
|
o pkg/mom_vecinv: Add energy & enstrophy conserving scheme (selectVortScheme=3) |
113 |
|
(from Sadourny, described by Burridge & Haseler, ECMWF Rep.4, 1977) |
114 |
|
o Refresh tutorial_global_oce_biogeo, notably: |
115 |
|
- change bathymetry (put back seamounts in ACC, use hFacs) |
116 |
|
- change EOS (POLY3 --> JMD95Z) |
117 |
|
- use Bryan-Lewis vertical diffusivity |
118 |
|
- use multidimensional advection |
119 |
|
- remove old virtual virtual in dic pkg, now done through ptracer pkg |
120 |
|
- change a couple of parameters in dic pkg |
121 |
|
o pkg/ctrl: change formula in adctrl_bound.F and add a parameter to ctrl_map_ini_gen.F |
122 |
|
o pkg/obcs: change default to OBCSfixTopo = .true., |
123 |
|
- set OBCSfixTopo=.false. in exp4/input/data.obcs to reproduce old result |
124 |
|
- replace "debugMode" with "debugLevel .GE. debLevB" |
125 |
|
o implementation of TAF directives for gchem,cfc,dic |
126 |
|
o remove "exp5" (replaced by tutorial_deep_convection). |
127 |
|
o pkg/obcs: add code to fix topography gradients normal to open boundaries; |
128 |
|
this requires computing the index fields OB_Jn/Js/Ie/Iw in obcs_readparms; |
129 |
|
the code can be turned on by setting OBCSfixTopo = .true. in data.obcs |
130 |
|
--> changes results in exp4, seaice_obcs |
131 |
|
o tutorial_deep_convection : |
132 |
|
update the set-up and increase resolution and domain size ; start |
133 |
|
from pseudo initial conditions = output at t=120mn in order to see |
134 |
|
something happening when running testreport. |
135 |
|
o pkg/seaice: add os7mp as an advection scheme |
136 |
|
add check for advection schemes |
137 |
|
add a flag for restoring T and S underneath sea ice |
138 |
|
o pkg/my82: add netcdf-support and diagnostics |
139 |
|
o thsice pkg: |
140 |
|
- Now hsMax limits snow height by turning it into ice following |
141 |
|
flooding scheme (and now conserving energy) |
142 |
|
- Slight rewriting of the formula in thsice_calc_thick.F |
143 |
|
- Parameter rhowi eliminated to be replaced by floodFac = (rhosw-rhoi)/rhos |
144 |
|
--> changes results of global_ocean.cs32x15.icedyn and global_ocean.cs32x15.thsice |
145 |
|
|
146 |
|
checkpoint59p |
147 |
|
o gchem / dic / cfc : |
148 |
|
- add useDIC, useCFC , read from "data.gchem" (& hold in GCHEM.h); |
149 |
|
- add new parameter files: "data.dic" & "data.cfc" ; |
150 |
|
- various re-arrangement in initialisation & header file ; |
151 |
|
- fix tutorial_global_oce_biogeo for multi-threaded test. |
152 |
|
o pkg/dic: test for "pickup_dic" file when restarting ; |
153 |
|
if no pickup-file, can restart with a re-initialised pH (same as it did |
154 |
|
before) providing pickupStrictlyMatch is set to False. |
155 |
|
o change argument list of S/R PTRACERS_WRITE_PICKUP & MYPACKAGE_WRITE_PICKUP |
156 |
|
+ fix call to S/R GGL90_WRITE_PICKUP & DIC_WRITE_PICKUP |
157 |
|
o add a "warning" (to Error msg file) when an Active Tracer is not using AB-2 |
158 |
|
and staggerTimeStep is off (needed for stability of Internal-wave dynamics). |
159 |
|
o move out of PARAMS.h & data the last pieces of old-ptracer code: tauTr1ClimRelax, |
160 |
|
& lambdaTr1ClimRelax ; put those 2 params in PTRACERS_PARAMS.h & data.ptracers |
161 |
|
(note: not clear to me what lambdaTr1ClimRelax is still used for ...) |
162 |
|
o pkg/mom_vecinv: - new parameter "selectVortScheme" to select which |
163 |
|
scheme to use for vorticty-term ; |
164 |
|
- compute upwindVorticity=T case within mom_vi_u/v_coriolis_c4.F |
165 |
|
(with default selectVortScheme, return to original discretisation) |
166 |
|
- remove weird combination (e.g.: upwindVorticity & highOrderVorticity) |
167 |
|
o pkg/seaice: add a line to the summary to indicate grid type (B/C) |
168 |
|
o in mom_calc_visc, fix average value of viscAh_W & viscA4_W next to the |
169 |
|
bottom ; only used with variable horizontal viscosity and non-hydrostatic. |
170 |
|
also fix diagnostics VISCAHW & VISCA4W. |
171 |
|
|
172 |
|
checkpoint59o |
173 |
|
o dome/code/obcs_calc.F: avoid division by zero (detected using gfortran). |
174 |
|
o change the loop ordering in most gad_*_adv_x.F to improve vectorization: |
175 |
|
move a few statemens out the main loop into separate loops |
176 |
|
o pkg/shelfice: |
177 |
|
- sort out freshwater flux; convert freshwater fluxes to mass fluxes |
178 |
|
(kg/m^2/s), does not break tests, because in verification/isomip, |
179 |
|
rhoConstFresh=rhoConst (by accident) |
180 |
|
- rearrange computation of tLoc, sLoc, pLoc to make code more efficient |
181 |
|
- move S/R find_rho_scalar, pkg/ecco/sw_ptmp.F, pkg/ecco/sw_adtg.F into |
182 |
|
new file model/src/seawater.F, so that they are available for all pkgs. |
183 |
|
- convert potential temperature into in-situ temperature. This changes the |
184 |
|
isomip tests => update tests |
185 |
|
o add a flag writePickupAtEnd (default=true) to be able to suppress |
186 |
|
writing a pickup at the last timestep |
187 |
|
o tools/mpack-1.6 : |
188 |
|
- disable automake (not working) and autoconf (not needed). |
189 |
|
- update to version 1.6.4 which fix MD5 code on some 64 bits platforms |
190 |
|
also cleaner with much less compilation warnings. |
191 |
|
o pkg/profiles: |
192 |
|
rename profiles_interp_mean_genericgrid as profiles_interp_mean_gg |
193 |
|
(for "less than 32 characters name" compiler requirement) |
194 |
|
o pkg/obcs and pkg/exf: added capability to specify seaice open boundary |
195 |
|
conditions that have a different sampling frequency than S/T/U/V obcs |
196 |
|
o pkg/diagnostics (stats-diags): |
197 |
|
- fix global Min/Max in case 1 processor get only empty tiles |
198 |
o genmake2: replace (4 times) |
o genmake2: replace (4 times) |
199 |
make genmake_tc_1.o >> genmake_warnings 2>&1 |
make genmake_tc_1.o >> genmake_warnings 2>&1 |
200 |
with |
with |