| 4 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
| 5 |
============================== |
============================== |
| 6 |
|
|
| 7 |
|
o autodiff |
| 8 |
|
Change default for TAF usage by removing argument '-nonew_arg' |
| 9 |
|
This changes TAF subroutine argument list (ordering of active var.) |
| 10 |
|
Omission of this flag leads to incompatibility w.r.t. TAMC |
| 11 |
|
To restore TAMC-compatibility, need following modifs: |
| 12 |
|
* use adoptfile tools/adjoint_options/adjoint_tamc_compatibility |
| 13 |
|
* use CPP option #define AUTODIFF_TAMC_COMPATIBILITY |
| 14 |
|
Tested TAF version is 1.9.22 |
| 15 |
|
N.B.: exch2 hand-written adjoint code currently not TAMC compatible |
| 16 |
|
|
| 17 |
|
checkpoint60 |
| 18 |
|
--->>> This is the last checkpoint for which generating the adjoint |
| 19 |
|
--->>> is done via the TAMC-compatibility TAF-flag "-nonewarg" |
| 20 |
|
--->>> by default. |
| 21 |
|
--->>> Future checkpoints will ommit this flag by default. |
| 22 |
|
o pkg/generic_advdiff, os7mp: simplify and avoid division by zero. |
| 23 |
|
o pkg/grdchk: Add gradient check to etan0 perturbations. |
| 24 |
|
o pkg/exf: read ice area from file (#ifdef ALLOW_ICE_AREAMASK). |
| 25 |
|
o pkg/autodiff: more flexibility in zeroadj. |
| 26 |
|
model/src: modifying do_oceanic_phys.F accordingly. |
| 27 |
|
o pkg/ctrl: Bounds for (kapgm,kapredi,eddypsi) parameter estimation. |
| 28 |
|
o model/src and pkg/gmredi: GMREDI_WITH_STABLE_ADJOINT CPP option |
| 29 |
|
special setup of gmredi for which the adjoint is stable |
| 30 |
|
enough for parameter optimization. |
| 31 |
|
o bridging the gap between eddy stress and GM. |
| 32 |
|
-> eddyTau is replaced with eddyPsi (eddyTau = f x rho0 x eddyPsi) |
| 33 |
|
along with a change in CPP option (now ALLOW_EDDYPSI). |
| 34 |
|
-> when using GM w/ GM_AdvForm: |
| 35 |
|
The total eddy streamfunction (Psi = eddyPsi + K x Slope) |
| 36 |
|
is applied either in the tracer Eq. or in momentum Eq. |
| 37 |
|
depending on data.gmredi (intro. GM_InMomAsStress). |
| 38 |
|
-> ALLOW_EDDYPSI_CONTROL for estimation purpose. |
| 39 |
|
The key modifications are in model/src/taueddy_external_forcing.F |
| 40 |
|
pkg/gmredi/gmredi_calc_*F pkg/gmredi/gmredi_*transport.F |
| 41 |
|
|
| 42 |
|
checkpoint59r |
| 43 |
|
o new S/R (pkg/mdsio/mdsio_facef_read.F) reading of sNx+1 x sNy+1 array |
| 44 |
|
from per-face file (take S/R READSYMTILE_RS out of ini_curvilinear_grid.F); |
| 45 |
|
fix option to read coriolis(@ cell-corner) from files with CS-grid. |
| 46 |
|
o pkg/mnc: add some attributes to standard 1D-coordinate variables |
| 47 |
|
o pkg/diagnostics: enforce missing_value attribute for scalar variables at |
| 48 |
|
mass points, when using mnc for output |
| 49 |
|
- add runtime parameter (missing_value, missing_value_int) to override |
| 50 |
|
default (which is the value of getcon('UNDEF')) |
| 51 |
|
o pkg/mnc: prepare mnc for the actual use of missing values with the |
| 52 |
|
diagnostics package: small fix in mnc_cw_readwrite.template and more |
| 53 |
|
initialisation of never used fields |
| 54 |
|
- replace mnc_common.h and mnc_id_header.h with corresponding upper case |
| 55 |
|
versions |
| 56 |
|
o pkg/thsice/thsice_ini_vars.F added default initial values so that |
| 57 |
|
thermodynamics sea ice solver does not blow up when some initial fields |
| 58 |
|
are missing. |
| 59 |
|
o move from external_forcing_surf.F to ini_forcing.F |
| 60 |
|
the p-coord. hack consisting in using pLoad to set phi0surf ; |
| 61 |
|
update output of tutorial_global_oce_in_p (affected @ truncation level). |
| 62 |
|
o Freshen up a bit hs94.1x64x5, notably: |
| 63 |
|
remove rotationPeriod and tRef=5*0. (adjoint only), and add |
| 64 |
|
exactConserv=.TRUE. |
| 65 |
|
o pkg/generic_advdiff: |
| 66 |
|
change option GAD_ALLOW_SOM_ADVECT to GAD_ALLOW_TS_SOM_ADV which only |
| 67 |
|
applies to files where Temperature & Salinity 2nd Order moments are used. |
| 68 |
|
o pkg/ptracers: add second-order moment advection schemes (80 and 81); |
| 69 |
|
this uses a dynamically allocated internal state data structure |
| 70 |
|
(#define PTRACERS_ALLOW_DYN_STATE in PTRACERS_OPTIONS.h) |
| 71 |
|
and requires a fortran 90 compiler |
| 72 |
|
|
| 73 |
|
checkpoint59q |
| 74 |
|
o pkg/mom_vecinv: Add energy & enstrophy conserving scheme (selectVortScheme=3) |
| 75 |
|
(from Sadourny, described by Burridge & Haseler, ECMWF Rep.4, 1977) |
| 76 |
|
o Refresh tutorial_global_oce_biogeo, notably: |
| 77 |
|
- change bathymetry (put back seamounts in ACC, use hFacs) |
| 78 |
|
- change EOS (POLY3 --> JMD95Z) |
| 79 |
|
- use Bryan-Lewis vertical diffusivity |
| 80 |
|
- use multidimensional advection |
| 81 |
|
- remove old virtual virtual in dic pkg, now done through ptracer pkg |
| 82 |
|
- change a couple of parameters in dic pkg |
| 83 |
|
o pkg/ctrl: change formula in adctrl_bound.F and add a parameter to ctrl_map_ini_gen.F |
| 84 |
|
o pkg/obcs: change default to OBCSfixTopo = .true., |
| 85 |
|
- set OBCSfixTopo=.false. in exp4/input/data.obcs to reproduce old result |
| 86 |
|
- replace "debugMode" with "debugLevel .GE. debLevB" |
| 87 |
|
o implementation of TAF directives for gchem,cfc,dic |
| 88 |
|
o remove "exp5" (replaced by tutorial_deep_convection). |
| 89 |
|
o pkg/obcs: add code to fix topography gradients normal to open boundaries; |
| 90 |
|
this requires computing the index fields OB_Jn/Js/Ie/Iw in obcs_readparms; |
| 91 |
|
the code can be turned on by setting OBCSfixTopo = .true. in data.obcs |
| 92 |
|
--> changes results in exp4, seaice_obcs |
| 93 |
|
o tutorial_deep_convection : |
| 94 |
|
update the set-up and increase resolution and domain size ; start |
| 95 |
|
from pseudo initial conditions = output at t=120mn in order to see |
| 96 |
|
something happening when running testreport. |
| 97 |
|
o pkg/seaice: add os7mp as an advection scheme |
| 98 |
|
add check for advection schemes |
| 99 |
|
add a flag for restoring T and S underneath sea ice |
| 100 |
|
o pkg/my82: add netcdf-support and diagnostics |
| 101 |
|
o thsice pkg: |
| 102 |
|
- Now hsMax limits snow height by turning it into ice following |
| 103 |
|
flooding scheme (and now conserving energy) |
| 104 |
|
- Slight rewriting of the formula in thsice_calc_thick.F |
| 105 |
|
- Parameter rhowi eliminated to be replaced by floodFac = (rhosw-rhoi)/rhos |
| 106 |
|
--> changes results of global_ocean.cs32x15.icedyn and global_ocean.cs32x15.thsice |
| 107 |
|
|
| 108 |
|
checkpoint59p |
| 109 |
|
o gchem / dic / cfc : |
| 110 |
|
- add useDIC, useCFC , read from "data.gchem" (& hold in GCHEM.h); |
| 111 |
|
- add new parameter files: "data.dic" & "data.cfc" ; |
| 112 |
|
- various re-arrangement in initialisation & header file ; |
| 113 |
|
- fix tutorial_global_oce_biogeo for multi-threaded test. |
| 114 |
|
o pkg/dic: test for "pickup_dic" file when restarting ; |
| 115 |
|
if no pickup-file, can restart with a re-initialised pH (same as it did |
| 116 |
|
before) providing pickupStrictlyMatch is set to False. |
| 117 |
|
o change argument list of S/R PTRACERS_WRITE_PICKUP & MYPACKAGE_WRITE_PICKUP |
| 118 |
|
+ fix call to S/R GGL90_WRITE_PICKUP & DIC_WRITE_PICKUP |
| 119 |
|
o add a "warning" (to Error msg file) when an Active Tracer is not using AB-2 |
| 120 |
|
and staggerTimeStep is off (needed for stability of Internal-wave dynamics). |
| 121 |
|
o move out of PARAMS.h & data the last pieces of old-ptracer code: tauTr1ClimRelax, |
| 122 |
|
& lambdaTr1ClimRelax ; put those 2 params in PTRACERS_PARAMS.h & data.ptracers |
| 123 |
|
(note: not clear to me what lambdaTr1ClimRelax is still used for ...) |
| 124 |
|
o pkg/mom_vecinv: - new parameter "selectVortScheme" to select which |
| 125 |
|
scheme to use for vorticty-term ; |
| 126 |
|
- compute upwindVorticity=T case within mom_vi_u/v_coriolis_c4.F |
| 127 |
|
(with default selectVortScheme, return to original discretisation) |
| 128 |
|
- remove weird combination (e.g.: upwindVorticity & highOrderVorticity) |
| 129 |
|
o pkg/seaice: add a line to the summary to indicate grid type (B/C) |
| 130 |
|
o in mom_calc_visc, fix average value of viscAh_W & viscA4_W next to the |
| 131 |
|
bottom ; only used with variable horizontal viscosity and non-hydrostatic. |
| 132 |
|
also fix diagnostics VISCAHW & VISCA4W. |
| 133 |
|
|
| 134 |
|
checkpoint59o |
| 135 |
o dome/code/obcs_calc.F: avoid division by zero (detected using gfortran). |
o dome/code/obcs_calc.F: avoid division by zero (detected using gfortran). |
| 136 |
o change the loop ordering in most gad_*_adv_x.F to improve vectorization: |
o change the loop ordering in most gad_*_adv_x.F to improve vectorization: |
| 137 |
move a few statemens out the main loop into separate loops |
move a few statemens out the main loop into separate loops |
| 146 |
isomip tests => update tests |
isomip tests => update tests |
| 147 |
o add a flag writePickupAtEnd (default=true) to be able to suppress |
o add a flag writePickupAtEnd (default=true) to be able to suppress |
| 148 |
writing a pickup at the last timestep |
writing a pickup at the last timestep |
| 149 |
|
o tools/mpack-1.6 : |
| 150 |
|
- disable automake (not working) and autoconf (not needed). |
| 151 |
|
- update to version 1.6.4 which fix MD5 code on some 64 bits platforms |
| 152 |
|
also cleaner with much less compilation warnings. |
| 153 |
o pkg/profiles: |
o pkg/profiles: |
| 154 |
rename profiles_interp_mean_genericgrid as profiles_interp_mean_gg |
rename profiles_interp_mean_genericgrid as profiles_interp_mean_gg |
| 155 |
(for "less than 32 characters name" compiler requirement) |
(for "less than 32 characters name" compiler requirement) |
Parent Directory
| 
Revision Log
| 
Revision Graph
| 
Patch