| 4 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
| 5 |
============================== |
============================== |
| 6 |
|
|
| 7 |
|
o store 3-D (in-situ) density in commom block (DYNVARS.h) |
| 8 |
|
- save 1 rho computation. |
| 9 |
|
- no longer modify T & S inland when using down_slope pkg. |
| 10 |
|
o gad_som : allow to switch to SOM advection for T & S without corresponding |
| 11 |
|
pickup files (using new S/R: MDS_CHECK4FILE). |
| 12 |
|
o keep removing _BEGIN/_END_MASTER between MDSIO S/R call |
| 13 |
|
(for useSingleCpuIO to work in multi-threaded) |
| 14 |
|
o pkg/shelfice: |
| 15 |
|
- disentangle parameters: retire shelfIceFile in data/PARM05 and substitute |
| 16 |
|
with SHELFICEtopoFile in data.shelfice |
| 17 |
|
- move code for reading ice shelf topography and modifying hFacC into separate |
| 18 |
|
subroutines of the shelfice package |
| 19 |
|
- separate initialisation of "fixed" and "variable" variables |
| 20 |
|
- kTopC is not (yet) replaced by kSurfC, although this is possible |
| 21 |
|
o fix broken diagnostic SIatmFW and changed the units from m/s to kg/m^2/s |
| 22 |
|
o add initialisation of deepFac if using Pcoords (new S/R set_grid_factors) |
| 23 |
|
and rename: |
| 24 |
|
ini_phiref.F --> set_ref_state.F (+ set anelastic factors) |
| 25 |
|
ini_reference_state.F --> load_ref_files.F (- set anelastic factors) |
| 26 |
|
o change FIND_RHO_SCALAR : return rho (instead of rho - rhoConst) |
| 27 |
|
|
| 28 |
|
checkpoint61c |
| 29 |
|
o add option ALLOW_ADDFLUID and run-time selector "selectAddFluid" |
| 30 |
|
to add mass source/sink of fluid in the interior (3-D generalisation |
| 31 |
|
of oceanic real-fresh water flux). |
| 32 |
|
o pkg/monitor: fix recip_dr argument in mon_advcflw2 call (-> advcfl_W_hf_max) |
| 33 |
|
o vermix test experiment: switch from LINEAR EOS to MDJWF (not tested before) |
| 34 |
|
o new pkg: down_slope |
| 35 |
|
density-driven downsloping flow parameterisation (Tellus, 1999, 412-430). |
| 36 |
|
- presently, modify T & S inland (dwnslp_calc_rho.F). |
| 37 |
|
can be avoided if rho was stored in common bloc and used in calc_phi_hyd. |
| 38 |
|
- implementation for ocean in P-coords not tested. |
| 39 |
|
- add secondary test to global_ocean.90x40x15 (input.dwnslp) to check this pkg. |
| 40 |
|
o move kSurfC,kSurfW & kSurfS from SURFACE.h to GRID.h (next to kLowC) |
| 41 |
|
(to facilitate conversion of oceanic pkg to p-coordinate). |
| 42 |
|
o change FIND_RHO to FIND_RHO_2D : |
| 43 |
|
pass arguments tFld & sFld as local (tile) 2D-arrays ; +fix EOS_CHECK. |
| 44 |
|
o new pkg: cheapaml |
| 45 |
|
|
| 46 |
|
checkpoint61b |
| 47 |
|
o exch2: |
| 48 |
|
change index bounds in rx2_cube exchanges (new S/R: EXCH2_GET_UV_BOUNDS) |
| 49 |
|
- no longer depend on the order sequence (N,S,E,W). |
| 50 |
|
- 3rd exchange no longer needed (tested with 24 tiles). |
| 51 |
|
- same modif to hand-written adjoint S/R (global_ocean.cs32x15: zero diff) |
| 52 |
|
- exch_UV_A-grid readily available (but not yet tested). |
| 53 |
|
comment out 3rd exch call and remove copy of V to U at SE & NE face-corners |
| 54 |
|
(both are no longer needed) in exch_uv: no effect in forward run; but |
| 55 |
|
get truncation error differences with adjoint exch. |
| 56 |
|
o exch2: |
| 57 |
|
- change index-bounds storage (moved from target to local tile, |
| 58 |
|
more intuitive this way) ; rename/remove some variables. |
| 59 |
|
- change utils/exch2/matlab-topology-generator & code_mods accordingly. |
| 60 |
|
|
| 61 |
|
checkpoint61a |
| 62 |
|
o exch2/matlab-topology-generator: |
| 63 |
|
fix bug in index bounds for N-S internal exchange (hard to tell if this |
| 64 |
|
had any effect on the solution). |
| 65 |
|
o pkg/mnc: small fix in mnc_handle_err: make sure that length of msgbuf |
| 66 |
|
is exceeded |
| 67 |
|
o testreport: modify syntax of awk, so that it works also with |
| 68 |
|
awk, and not only with nawk and gawk |
| 69 |
|
|
| 70 |
|
checkpoint61 |
| 71 |
|
o autodiff |
| 72 |
|
Change default for TAF usage by removing argument '-nonew_arg' |
| 73 |
|
This changes TAF subroutine argument list (ordering of active var.) |
| 74 |
|
Omission of this flag leads to incompatibility w.r.t. TAMC |
| 75 |
|
To restore TAMC-compatibility, need following modifs: |
| 76 |
|
* use adoptfile tools/adjoint_options/adjoint_tamc_compatibility |
| 77 |
|
* use CPP option #define AUTODIFF_TAMC_COMPATIBILITY |
| 78 |
|
Tested TAF version is 1.9.22 |
| 79 |
|
N.B.: exch2 hand-written adjoint code currently not TAMC compatible |
| 80 |
|
|
| 81 |
|
checkpoint60 |
| 82 |
|
--->>> This is the last checkpoint for which generating the adjoint |
| 83 |
|
--->>> is done via the TAMC-compatibility TAF-flag "-nonewarg" |
| 84 |
|
--->>> by default. |
| 85 |
|
--->>> Future checkpoints will ommit this flag by default. |
| 86 |
|
o pkg/generic_advdiff, os7mp: simplify and avoid division by zero. |
| 87 |
|
o pkg/grdchk: Add gradient check to etan0 perturbations. |
| 88 |
|
o pkg/exf: read ice area from file (#ifdef ALLOW_ICE_AREAMASK). |
| 89 |
|
o pkg/autodiff: more flexibility in zeroadj. |
| 90 |
|
model/src: modifying do_oceanic_phys.F accordingly. |
| 91 |
|
o pkg/ctrl: Bounds for (kapgm,kapredi,eddypsi) parameter estimation. |
| 92 |
|
o model/src and pkg/gmredi: GMREDI_WITH_STABLE_ADJOINT CPP option |
| 93 |
|
special setup of gmredi for which the adjoint is stable |
| 94 |
|
enough for parameter optimization. |
| 95 |
|
o bridging the gap between eddy stress and GM. |
| 96 |
|
-> eddyTau is replaced with eddyPsi (eddyTau = f x rho0 x eddyPsi) |
| 97 |
|
along with a change in CPP option (now ALLOW_EDDYPSI). |
| 98 |
|
-> when using GM w/ GM_AdvForm: |
| 99 |
|
The total eddy streamfunction (Psi = eddyPsi + K x Slope) |
| 100 |
|
is applied either in the tracer Eq. or in momentum Eq. |
| 101 |
|
depending on data.gmredi (intro. GM_InMomAsStress). |
| 102 |
|
-> ALLOW_EDDYPSI_CONTROL for estimation purpose. |
| 103 |
|
The key modifications are in model/src/taueddy_external_forcing.F |
| 104 |
|
pkg/gmredi/gmredi_calc_*F pkg/gmredi/gmredi_*transport.F |
| 105 |
|
|
| 106 |
|
checkpoint59r |
| 107 |
|
o new S/R (pkg/mdsio/mdsio_facef_read.F) reading of sNx+1 x sNy+1 array |
| 108 |
|
from per-face file (take S/R READSYMTILE_RS out of ini_curvilinear_grid.F); |
| 109 |
|
fix option to read coriolis(@ cell-corner) from files with CS-grid. |
| 110 |
|
o pkg/mnc: add some attributes to standard 1D-coordinate variables |
| 111 |
|
o pkg/diagnostics: enforce missing_value attribute for scalar variables at |
| 112 |
|
mass points, when using mnc for output |
| 113 |
|
- add runtime parameter (missing_value, missing_value_int) to override |
| 114 |
|
default (which is the value of getcon('UNDEF')) |
| 115 |
|
o pkg/mnc: prepare mnc for the actual use of missing values with the |
| 116 |
|
diagnostics package: small fix in mnc_cw_readwrite.template and more |
| 117 |
|
initialisation of never used fields |
| 118 |
|
- replace mnc_common.h and mnc_id_header.h with corresponding upper case |
| 119 |
|
versions |
| 120 |
|
o pkg/thsice/thsice_ini_vars.F added default initial values so that |
| 121 |
|
thermodynamics sea ice solver does not blow up when some initial fields |
| 122 |
|
are missing. |
| 123 |
|
o move from external_forcing_surf.F to ini_forcing.F |
| 124 |
|
the p-coord. hack consisting in using pLoad to set phi0surf ; |
| 125 |
|
update output of tutorial_global_oce_in_p (affected @ truncation level). |
| 126 |
|
o Freshen up a bit hs94.1x64x5, notably: |
| 127 |
|
remove rotationPeriod and tRef=5*0. (adjoint only), and add |
| 128 |
|
exactConserv=.TRUE. |
| 129 |
|
o pkg/generic_advdiff: |
| 130 |
|
change option GAD_ALLOW_SOM_ADVECT to GAD_ALLOW_TS_SOM_ADV which only |
| 131 |
|
applies to files where Temperature & Salinity 2nd Order moments are used. |
| 132 |
|
o pkg/ptracers: add second-order moment advection schemes (80 and 81); |
| 133 |
|
this uses a dynamically allocated internal state data structure |
| 134 |
|
(#define PTRACERS_ALLOW_DYN_STATE in PTRACERS_OPTIONS.h) |
| 135 |
|
and requires a fortran 90 compiler |
| 136 |
|
|
| 137 |
|
checkpoint59q |
| 138 |
|
o pkg/mom_vecinv: Add energy & enstrophy conserving scheme (selectVortScheme=3) |
| 139 |
|
(from Sadourny, described by Burridge & Haseler, ECMWF Rep.4, 1977) |
| 140 |
|
o Refresh tutorial_global_oce_biogeo, notably: |
| 141 |
|
- change bathymetry (put back seamounts in ACC, use hFacs) |
| 142 |
|
- change EOS (POLY3 --> JMD95Z) |
| 143 |
|
- use Bryan-Lewis vertical diffusivity |
| 144 |
|
- use multidimensional advection |
| 145 |
|
- remove old virtual virtual in dic pkg, now done through ptracer pkg |
| 146 |
|
- change a couple of parameters in dic pkg |
| 147 |
|
o pkg/ctrl: change formula in adctrl_bound.F and add a parameter to ctrl_map_ini_gen.F |
| 148 |
|
o pkg/obcs: change default to OBCSfixTopo = .true., |
| 149 |
|
- set OBCSfixTopo=.false. in exp4/input/data.obcs to reproduce old result |
| 150 |
|
- replace "debugMode" with "debugLevel .GE. debLevB" |
| 151 |
|
o implementation of TAF directives for gchem,cfc,dic |
| 152 |
|
o remove "exp5" (replaced by tutorial_deep_convection). |
| 153 |
|
o pkg/obcs: add code to fix topography gradients normal to open boundaries; |
| 154 |
|
this requires computing the index fields OB_Jn/Js/Ie/Iw in obcs_readparms; |
| 155 |
|
the code can be turned on by setting OBCSfixTopo = .true. in data.obcs |
| 156 |
|
--> changes results in exp4, seaice_obcs |
| 157 |
|
o tutorial_deep_convection : |
| 158 |
|
update the set-up and increase resolution and domain size ; start |
| 159 |
|
from pseudo initial conditions = output at t=120mn in order to see |
| 160 |
|
something happening when running testreport. |
| 161 |
|
o pkg/seaice: add os7mp as an advection scheme |
| 162 |
|
add check for advection schemes |
| 163 |
|
add a flag for restoring T and S underneath sea ice |
| 164 |
|
o pkg/my82: add netcdf-support and diagnostics |
| 165 |
|
o thsice pkg: |
| 166 |
|
- Now hsMax limits snow height by turning it into ice following |
| 167 |
|
flooding scheme (and now conserving energy) |
| 168 |
|
- Slight rewriting of the formula in thsice_calc_thick.F |
| 169 |
|
- Parameter rhowi eliminated to be replaced by floodFac = (rhosw-rhoi)/rhos |
| 170 |
|
--> changes results of global_ocean.cs32x15.icedyn and global_ocean.cs32x15.thsice |
| 171 |
|
|
| 172 |
|
checkpoint59p |
| 173 |
|
o gchem / dic / cfc : |
| 174 |
|
- add useDIC, useCFC , read from "data.gchem" (& hold in GCHEM.h); |
| 175 |
|
- add new parameter files: "data.dic" & "data.cfc" ; |
| 176 |
|
- various re-arrangement in initialisation & header file ; |
| 177 |
|
- fix tutorial_global_oce_biogeo for multi-threaded test. |
| 178 |
|
o pkg/dic: test for "pickup_dic" file when restarting ; |
| 179 |
|
if no pickup-file, can restart with a re-initialised pH (same as it did |
| 180 |
|
before) providing pickupStrictlyMatch is set to False. |
| 181 |
|
o change argument list of S/R PTRACERS_WRITE_PICKUP & MYPACKAGE_WRITE_PICKUP |
| 182 |
|
+ fix call to S/R GGL90_WRITE_PICKUP & DIC_WRITE_PICKUP |
| 183 |
|
o add a "warning" (to Error msg file) when an Active Tracer is not using AB-2 |
| 184 |
|
and staggerTimeStep is off (needed for stability of Internal-wave dynamics). |
| 185 |
|
o move out of PARAMS.h & data the last pieces of old-ptracer code: tauTr1ClimRelax, |
| 186 |
|
& lambdaTr1ClimRelax ; put those 2 params in PTRACERS_PARAMS.h & data.ptracers |
| 187 |
|
(note: not clear to me what lambdaTr1ClimRelax is still used for ...) |
| 188 |
|
o pkg/mom_vecinv: - new parameter "selectVortScheme" to select which |
| 189 |
|
scheme to use for vorticty-term ; |
| 190 |
|
- compute upwindVorticity=T case within mom_vi_u/v_coriolis_c4.F |
| 191 |
|
(with default selectVortScheme, return to original discretisation) |
| 192 |
|
- remove weird combination (e.g.: upwindVorticity & highOrderVorticity) |
| 193 |
|
o pkg/seaice: add a line to the summary to indicate grid type (B/C) |
| 194 |
|
o in mom_calc_visc, fix average value of viscAh_W & viscA4_W next to the |
| 195 |
|
bottom ; only used with variable horizontal viscosity and non-hydrostatic. |
| 196 |
|
also fix diagnostics VISCAHW & VISCA4W. |
| 197 |
|
|
| 198 |
|
checkpoint59o |
| 199 |
|
o dome/code/obcs_calc.F: avoid division by zero (detected using gfortran). |
| 200 |
|
o change the loop ordering in most gad_*_adv_x.F to improve vectorization: |
| 201 |
|
move a few statemens out the main loop into separate loops |
| 202 |
|
o pkg/shelfice: |
| 203 |
|
- sort out freshwater flux; convert freshwater fluxes to mass fluxes |
| 204 |
|
(kg/m^2/s), does not break tests, because in verification/isomip, |
| 205 |
|
rhoConstFresh=rhoConst (by accident) |
| 206 |
|
- rearrange computation of tLoc, sLoc, pLoc to make code more efficient |
| 207 |
|
- move S/R find_rho_scalar, pkg/ecco/sw_ptmp.F, pkg/ecco/sw_adtg.F into |
| 208 |
|
new file model/src/seawater.F, so that they are available for all pkgs. |
| 209 |
|
- convert potential temperature into in-situ temperature. This changes the |
| 210 |
|
isomip tests => update tests |
| 211 |
|
o add a flag writePickupAtEnd (default=true) to be able to suppress |
| 212 |
|
writing a pickup at the last timestep |
| 213 |
|
o tools/mpack-1.6 : |
| 214 |
|
- disable automake (not working) and autoconf (not needed). |
| 215 |
|
- update to version 1.6.4 which fix MD5 code on some 64 bits platforms |
| 216 |
|
also cleaner with much less compilation warnings. |
| 217 |
|
o pkg/profiles: |
| 218 |
|
rename profiles_interp_mean_genericgrid as profiles_interp_mean_gg |
| 219 |
|
(for "less than 32 characters name" compiler requirement) |
| 220 |
|
o pkg/obcs and pkg/exf: added capability to specify seaice open boundary |
| 221 |
|
conditions that have a different sampling frequency than S/T/U/V obcs |
| 222 |
|
o pkg/diagnostics (stats-diags): |
| 223 |
|
- fix global Min/Max in case 1 processor get only empty tiles |
| 224 |
|
o genmake2: replace (4 times) |
| 225 |
|
make genmake_tc_1.o >> genmake_warnings 2>&1 |
| 226 |
|
with |
| 227 |
|
COMM="$CC $CFLAGS -c genmake_tc_1.c" |
| 228 |
|
echo $COMM >> genmake_warnings |
| 229 |
|
$COMM >> genmake_warnings 2>&1 |
| 230 |
|
to make genmake2 always see the value of CC as specified in the optfile. |
| 231 |
|
o generic_advdiff: - prather advection scheme (SOM) coded for CS-grid. |
| 232 |
|
- add diagnostics for 1srt & 2nd Order moments and sub-grid variance. |
| 233 |
|
- modify verification advect_cs test to test SOM-advection on CS-grid. |
| 234 |
|
o pkg/ctrl: |
| 235 |
|
- Rewrite of ctrl_map_ini_ecco.F, |
| 236 |
|
introducing a generic routine (ctrl_map_ini_gen.F). |
| 237 |
|
- Modification of ctrl_bound/adctrl_bound handling control vector bounds. |
| 238 |
|
|
| 239 |
|
checkpoint59n |
| 240 |
|
o pkg/diagnostics: for large number of diagnostics, store diagnostic mate |
| 241 |
|
number in dedicate array "hdiag": |
| 242 |
|
- new version of S/R DIAGNOSTICS_ADD2LIST : DIAGNOSTICS_ADDTOLIST |
| 243 |
|
with 1 more argument (mate number); modify old version DIAGNOSTICS_ADD2LIST. |
| 244 |
|
- many small changes (integer format) for large number of diagnostics |
| 245 |
|
- start to use DIAGNOSTICS_ADDTOLIST (necessary for diagnostics with |
| 246 |
|
a counter mate) in pkg diagnostics, ptracers, thsice & aim_v23. |
| 247 |
|
o add parameter rotateGrid and Euler angles EulerPhi/Theta/Psi. For |
| 248 |
|
usingSphericalGrid, this allows to define the rotated grid coordinates |
| 249 |
|
via phiMin/thetaMin/dxSpacing/dySpacing, etc., but then re-compute the |
| 250 |
|
geographical coordinates according to the inverse of the rotation defined |
| 251 |
|
by the Euler angles, so that Coriolis parameter and online-interpolation by |
| 252 |
|
exf works with the geographical coordinates on XC/YC/XG/YG. |
| 253 |
|
Naturally, this feature does not work with all packages, so some |
| 254 |
|
combinations are prohibited in config_summary (flt, flt_zonal, ecco, |
| 255 |
|
profiles), because there the coordinates are assumed to be regular |
| 256 |
|
spherical grid coordinates. |
| 257 |
|
o pkg/gmredi,ctrl,ecco,autodiff,grdchk and model/src,inc: |
| 258 |
|
- introduce isopycnal diffusion coefficient control. |
| 259 |
|
o tools/adjoint_options: |
| 260 |
|
- additional control vector components. |
| 261 |
|
o pkg/mnc: have mnc create marginally meaning full coordinate variables |
| 262 |
|
(that is the grid indices) if usingCurvilinearGrid and NOT EXCH2 |
| 263 |
|
o pkg/mypackage: |
| 264 |
|
- add example of read/write pickup for state-var ; |
| 265 |
|
- add entry points to the main code for state-vars diags & write-pickup. |
| 266 |
|
o pkg/exf: rewrite of exf_getffieldrec.F to properly deal with year |
| 267 |
|
transitions for useExfYearlyFields |
| 268 |
|
- add new subroutine that determines the file to read from for |
| 269 |
|
use*YearlyFields = .TRUE. and .FALSE. |
| 270 |
|
- add a verification experiment to global_with_exf that tests |
| 271 |
|
useExfYearlyFields with USE_EXF_INTERPOLATION defined |
| 272 |
|
o pkg/obcs/exf: add useOBCSYearlyFields (=.FALSE.) to obcs with exf in |
| 273 |
|
analogy to useExfYearlyFields; rearrange obcs_prescribe_read: two new |
| 274 |
|
subroutines (which could be generated from a template, because they are |
| 275 |
|
almost identical), for now they are included in obcs_prescribe_read.F for |
| 276 |
|
simplicity |
| 277 |
|
o pkg/exf: |
| 278 |
|
- add a little more diagnostic to exf_interp_read (in analogy to |
| 279 |
|
mds_read_field, can be turned off with debugLevel < debugLevA=1) |
| 280 |
|
- improve vectorizability of exf_interp for TARGET_NEC_SX, makes code |
| 281 |
|
ugly but fast by unrolling short inner loops, inserting a unroll=8 |
| 282 |
|
directive in routine lagran, and splitting a loop and making do while the |
| 283 |
|
outermost loop (the latter requires the intoduction of index arrays) |
| 284 |
|
o pkg/obcs: replace obcs_apply_area/heff/hsalt/hsnow with obcs_apply_seaice |
| 285 |
|
o pkg/seaice: moved DIFFERENT_MULTIPLE from seaice_model to seaice_dynsolver |
| 286 |
|
http://forge.csail.mit.edu/pipermail/mitgcm-devel/2008-January/003173.html |
| 287 |
|
o verification/seaice_obcs: updated for restart test and latest options |
| 288 |
|
o pkg/seaice: add a new flag SEAICEuseTEM to use a slightly different |
| 289 |
|
ice rheology (so-called truncated ellipse, that gets rid of tensile stress); |
| 290 |
|
for now this is only available for the C-grid code and also within |
| 291 |
|
#ifdef SEAICE_ALLOW_TEM, because this option may slow down the EVP code |
| 292 |
|
|
| 293 |
|
checkpoint59m |
| 294 |
|
o pkg/seaice: use different time constant for melting (SEAICE_gamma_t) |
| 295 |
|
and freezing (SEAICE_gamma_t_frz) ice due to mixed layer temperature |
| 296 |
|
o gad_som : fix flux output (& diagnostics) to have same units as other |
| 297 |
|
advection scheme fluxes (was previously multiplied by deltaT). |
| 298 |
|
o added verification/seaice_obcs both to test seaice open boundaries |
| 299 |
|
and to provide an example matlab scrpt for generating them |
| 300 |
|
o added verification/lab_sea/input.salt_plume for testing pkg/salt_plume |
| 301 |
|
o pkg/salt_plume: clean up salt_plume_frac.F to |
| 302 |
|
remove redundancy in PlumeMethod 1 and 4 |
| 303 |
|
o pkg/kpp: added saltplume diagnostics |
| 304 |
|
o pkg/obcs and pkg/seaice: |
| 305 |
|
- do not set uice and vice to zero when OB*icefile are unspecified |
| 306 |
|
- OBCS_SEAICE_AVOID_CONVERGENCE is undefined by default |
| 307 |
|
- moved seaice tracer obcs after call to seaice_growth |
| 308 |
|
o pkg/salt_plume: added various options for salt plume depth |
| 309 |
|
criterion and for vertical distribution of the salt |
| 310 |
|
|
| 311 |
|
checkpoint59l |
| 312 |
|
o pkg/ptracers: new version of ptracers pickup: |
| 313 |
|
read meta file and write only fields which are needed to restart. |
| 314 |
|
( same logic as main pickup file ; also using pickupStrictlyMatch ) |
| 315 |
|
o pkg/seaice: |
| 316 |
|
- add timescale for melting due to warm mixed layer as a means of |
| 317 |
|
setting SEAICE_availHeatFrac more conveniently |
| 318 |
|
- add a CPP-flag/runtime parameter check for flooding algorithm |
| 319 |
|
- reformulated availHeat in seaice_growth.F to be independent of AREA |
| 320 |
|
o change units attribute of EmPmR netcdf variables to kg/m^2/s to reflect recent |
| 321 |
|
unit changes |
| 322 |
|
o pkg/obcs: added OBCS_SEAICE_AVOID_CONVERGENCE constraint |
| 323 |
|
o cg2d.F: add 3 compiler directives that speed up this routine by 30% on a NEC |
| 324 |
|
SX vector computer. |
| 325 |
|
o pkg/seaice: |
| 326 |
|
- removed SEAICE_FFIELDS.h and seaice_get_forcing.F |
| 327 |
|
seaice forcing fields can now be read only through pkg/exf |
| 328 |
|
- removed some unused halos |
| 329 |
|
o no longer write pickup from THE_MODEL_MAIN but always from FORWARD_STEP |
| 330 |
|
o pkg/exf: added useRelativeWind EXF_PARAM.h option for subtracting U/VVEL |
| 331 |
|
or U/VICE from U/VWIND before computing U/VSTRESS |
| 332 |
|
|
| 333 |
|
checkpoint59k |
| 334 |
|
o pkg/seaice: ICE2WATR=SEAICE_rhoIce/rhoConst for consistency between |
| 335 |
|
dynamics and thermodynamics. Set ICE2WATR=0.92 to reproduce older results. |
| 336 |
|
o fix un-initialised variable problem (in mom_vecinv & shap_filt): |
| 337 |
|
all experiments pass when compile with un-initialised var. checking option. |
| 338 |
|
o added pkg/salt_plume |
| 339 |
|
o script to test restart (for now, in MITgcm_contrib/jmc_script) |
| 340 |
|
o pkg/seaice: new read/write pickup version. |
| 341 |
|
read meta file and write only fields which are needed to restart. |
| 342 |
|
( same logic as main pickup file ; also using pickupStrictlyMatch ) |
| 343 |
|
o pkg/seaice: fix bug in seaice_ocean_stress (call of seaice_calc_strainrates) |
| 344 |
|
o lab_sea.hb87: modify the domain & update output.hb87.txt : |
| 345 |
|
lat-long grid & no-slip BC in seaice-dynamics produce different sigma12 |
| 346 |
|
at j=1 & j=Ny+1 : this case is not treated correctly in the restart |
| 347 |
|
process. Fixed here by closing the domain at the Northern boundary. |
| 348 |
|
o pkg/rw: cleaning-up: call directly new MDSIO S/R MDS_READ/WRITE_FIELD. |
| 349 |
|
o pkg/mdsio: add new arguments to MDS_READ/WRITE_FIELD for reading/writing |
| 350 |
|
a set of consecutive levels in/from a 3-D array. |
| 351 |
|
o pkg/ptracers: |
| 352 |
|
use a 2 characters string ioLabel to identify tracer (if more than 99 tracers). |
| 353 |
|
to-do: fix pkg/diagnostics when large number of diagnostics are defined. |
| 354 |
|
|
| 355 |
|
checkpoint59j |
| 356 |
|
o in pkgs: dic, obcs, offline, rbcs: |
| 357 |
|
standard way to compute coeff. for time interpolation between 2 time reccords |
| 358 |
|
which retains full precision (& make results less dependent on compiler). |
| 359 |
|
update output of exp: tutorial_cfc_offline & tutorial_global_oce_biogeo. |
| 360 |
|
o pkg/cfc: if compiled with exf and seaice, cfc now can use fields from those |
| 361 |
|
packages instead of reading ice cover and wind speed separately |
| 362 |
|
o split PTRACERS.h in 2 header files: PTRACERS_FIELDS.h & PTRACERS_PARAMS.h |
| 363 |
|
(keep a temporary version of PTRACERS.h in verification/OpenAD/code_ad_openad |
| 364 |
|
to allow to compile customized code before the update) |
| 365 |
|
o cleaning-up: |
| 366 |
|
- fix CPP instruction syntax in pkg/obcs |
| 367 |
|
- replace (some) call to MDSIO S/R with call to RW interface S/R ; and |
| 368 |
|
add #ifdef ALLOW_MDSIO elsewhere (allows to compile without pkg/mdsio) |
| 369 |
|
- fix some threads issue in IO parts. |
| 370 |
|
o pkg/dic: |
| 371 |
|
- add tons of "_d 0" (--> much better testreports !) |
| 372 |
|
and then update outputs of tutorial_global_oce_biogeo |
| 373 |
|
o pkg/mom_fluxform: |
| 374 |
|
add a special version of momentum advection intended to conserve momentum |
| 375 |
|
next to a bathymetry step or a coastline edge. (default = off = #undef) |
| 376 |
|
o new read/write pickup version: |
| 377 |
|
- for now, only implemented for main (mdsio) pickup file. |
| 378 |
|
- write out only what is needed to restart; merge pickup_nh , pickup_ph |
| 379 |
|
and main pickup together. |
| 380 |
|
- read meta file; read only what is needed to restart; stop if a field |
| 381 |
|
is missing (if pickupStrictlyMatch=T, = the default) or try to restart |
| 382 |
|
if it make sense (even if the restart is not perfect). |
| 383 |
|
- if no meta file, assume that pickup to read corresponds exactly to |
| 384 |
|
a pickup that would currently be written. |
| 385 |
|
- read pickup written in the old format (until checkpoint59i) if it |
| 386 |
|
recognizes an old meta file (no field-list). |
| 387 |
|
- add a meta file in exp. which starts from a pickup: 2 cases: |
| 388 |
|
a) minimal meta file (with just file precision) + pickupStrictlyMatch=F |
| 389 |
|
e.g.: global_ocean.90x40x15/input/pickup.0000036000.meta |
| 390 |
|
b) a meta file describing the content of the associated pickup file |
| 391 |
|
e.g.: tutorial_global_oce_biogeo/input/pickup.0004248000.meta |
| 392 |
|
o pkg/mdsio & pkg/rw: |
| 393 |
|
- new routine to read a meta file ; new set of routine to read |
| 394 |
|
Multiple-fields file (with meta file). |
| 395 |
|
o added open boundary conditions capability for seaice |
| 396 |
|
HSNOW, HSALT, UICE, and VICE |
| 397 |
|
|
| 398 |
|
checkpoint59i |
| 399 |
|
o adjoint: |
| 400 |
|
* fix multiDimAdvection adjoint |
| 401 |
|
(partial recomputation problem by TAF for fVerT(:,:,kDown) |
| 402 |
|
and comlev1 size definition (maxpass vs. max. number of tracers) |
| 403 |
|
* verification/global_ocean.90x45x15 adjoint now tests |
| 404 |
|
NLFS and multiDimAdvection using DST3 (=30) |
| 405 |
|
* verification/lab_sea now tests multiDimAdvection using DST3 (=30) |
| 406 |
|
o pkg/kpp: add "_d 0" to real constant (if missing); |
| 407 |
|
update output of exp: natl_box (10.digits); vermix (std, 7.digits) |
| 408 |
|
and lab_sea, 3.FW (only 6,4,6 digits match) & AD (8.digits). |
| 409 |
|
o prepare for "clever pickup" implementation: add new header file "RESTART.h" |
| 410 |
|
for internal parameters related to restart process (formerly from PARAMS.h) |
| 411 |
|
and add startAB parameter to argument list of S/R ADAMS_BASHFORTH2. |
| 412 |
|
o more standard interface to {PKG}_WRITE_PICKUP with pickup suffix passed |
| 413 |
|
as argument. (only atm2d & obcs write_pickup have not yet been updated) |
| 414 |
|
o pkg/monitor: |
| 415 |
|
- use GLOBAL_SUM_TILE instead of _GLOBAL_SUM |
| 416 |
|
- fix mon_stats_rs (statistics over the full domain, even where field is zero) |
| 417 |
|
- account for deep Atmosphere factor and density factor (anelastic) |
| 418 |
|
o added open boundary conditions capability for seaice HEFF and AREA |
| 419 |
|
o pkg/shelfice: add option for convservative form of Hellmer&Olbers (1989) |
| 420 |
|
thermodynamics |
| 421 |
|
- requires a little reorganization, which affects the testreport results |
| 422 |
|
of isomip.htd (only 7 digits of cg2d agree) => update/modify experiment, |
| 423 |
|
so that now the conservative form is tested |
| 424 |
|
- add a few comments |
| 425 |
|
o add missing cvs Header: or Name: in eesupp/*, pkg/* & verification/code* |
| 426 |
|
o add a little bit of code to gad_os7mp_adv_?.F to help vectorization |
| 427 |
|
o add new diagnostics to seaice: advective, diffusive fluxes (only for |
| 428 |
|
multidim-advection, for scheme 2 they are not filled) and transports, |
| 429 |
|
thermodynamic growth rates |
| 430 |
|
o change units of EmPmR (inside the code, + within coupling interface) |
| 431 |
|
from m/s to kg/m2/s (mass flux) but leave input file unchanged. |
| 432 |
|
Affects results @ truncation level (cg2d match) and update output of |
| 433 |
|
experiments: lab_sea(6), global_ocean.90x40x15(10), |
| 434 |
|
global_ocean.cs32x15(11x3,10), natl_box(11), tutorial_global_oce_latlon(11) |
| 435 |
|
+ adjoint results: global_ocean_ebm(11) and lab_sea (12). |
| 436 |
|
Coupled set-up: allow to restart from previous pickups (cpl_oldPickup=T). |
| 437 |
|
Next: will retire convertEmP2rUnit (simply using mass2rUnit instead). |
| 438 |
|
|
| 439 |
|
checkpoint59h |
| 440 |
|
o pkg/exf: |
| 441 |
|
- improve vectorization by moving if-statement out of loops in |
| 442 |
|
exf_mapfields, exf_radiation |
| 443 |
|
- exf_bulkformulae.F: introduce 2D fields and vectorize code |
| 444 |
|
o pkg/kpp: added the salt plume scheme (ALLOW_SALT_PLUME) to KPP package |
| 445 |
|
o testreport: |
| 446 |
|
- allow to test for Qnet & SeaIce monitor output; clean-up ALLOW_BULK_OFFLINE |
| 447 |
|
monitor & update offline_exf_seaice exp. output. |
| 448 |
|
- AD testing: pass/FAIL according to Grad. output accuracy (instead of Cost) |
| 449 |
|
o pkg/seaice: |
| 450 |
|
- bug fix in the evp solver: boundary conditions were affected be |
| 451 |
|
masking of zMax/zMin (changes results) |
| 452 |
|
- minor bug fix in boundary conditions in seaice_calc_strainrates |
| 453 |
|
(changes results for no_slip) |
| 454 |
|
- seaice_lsr.F: make interpolation of eta dependent on boundaries |
| 455 |
|
- replace hardcoded lower limit for zeta (zMin) by runtime parameter |
| 456 |
|
- add a test to lab_sea, that tests the Hibler and Bryan (1987) ocean |
| 457 |
|
stress formulation and no-slip boundary conditions |
| 458 |
|
o pkg/ptracers: Temporary fix for PTRACERS_useGMRedi & PTRACERS_useKPP |
| 459 |
|
in gad_calc_rhs.F ; add few safety stops in initialisation phase. |
| 460 |
|
o pkg/seaice: a first HSALT time stepping equation has been checked in |
| 461 |
|
o pkg/seaice: replaced maskW with seaiceMaskU and maskS with seaiceMaskV |
| 462 |
|
in seaice_lsr.F; this solves the pkg/seaice-pkg/obcs problem |
| 463 |
|
without requiring DO-IF statements |
| 464 |
|
o pkg/thsice: |
| 465 |
|
use continuous power-law function for partition of energy between lateral |
| 466 |
|
melting/freezing and thinning/thickening ; old code using step function |
| 467 |
|
still available with #undef THSICE_FRACEN_POWERLAW |
| 468 |
|
update output of exp: aim.5l_cs.thSI, global_ocean.cs32x15.icedyn & .thsice |
| 469 |
|
o testreport: change AD testing : check against results/output_adm.txt |
| 470 |
|
using same scripts as for forward test. |
| 471 |
|
|
| 472 |
|
checkpoint59g |
| 473 |
|
o adjoint testing: in preparation for changing testreport: |
| 474 |
|
- output format: cost and grad written on separated lines |
| 475 |
|
- ad output results in directory "results" (instead of "results_ad") |
| 476 |
|
o pkg/seaice: adding ice salinity HSALT as a prognostic variable |
| 477 |
|
o pkg/seaice: changes for using pkg/seaice with pkg/obcs |
| 478 |
|
o pkg/ecco: allow user defined cost function contributions |
| 479 |
|
o testreport: |
| 480 |
|
Allow to choose (per experiment) the list of tested output variables |
| 481 |
|
and the main variable (used to issue a pass/FAIL ) ; |
| 482 |
|
- for now, done by adding the specific list in file: [exp]/input/tr_checklist |
| 483 |
|
- default is unchanged: >cg2d< , (T,S,U,V,5 tracers)_[min,max,mean,StD] ; |
| 484 |
|
Also modify printing of summary: |
| 485 |
|
a) ">" NN "<" indicates the "main" variable (used for pass/FAIL) |
| 486 |
|
b) echo the list of tested variables in the individual summary.txt of |
| 487 |
|
each experiment. |
| 488 |
|
o new S/R GLOBAL_SUM_TILE to compute global sum independently of number of |
| 489 |
|
threads and number of processes (with #define GLOBAL_SUM_SEND_RECV) |
| 490 |
|
(but still depend on the tile decompostion of the domain). |
| 491 |
|
o pkg/seaice: change a discretization detail to make the B-grid code |
| 492 |
|
"more symmetric". This change is likely to remove all "ridges/leads" that |
| 493 |
|
where quasi-aligned with the grid in B-grid solutions. It may also be |
| 494 |
|
the solution to the stress coupling issues with the B-grid. |
| 495 |
|
o each tested exp: add a "run" directory with a standard ".cvsignore" file ; |
| 496 |
|
remove .cvsignore from input dir. (since testreport no longer run here). |
| 497 |
|
o pkg/thsice: |
| 498 |
|
- cleaning: move common block out of "THSICE_2DYN.h" (removed) to "THSICE_VARS.h" |
| 499 |
|
- fix reshaping of sea-ice just after advection: |
| 500 |
|
melt all the ice only if ice-volume is too small. |
| 501 |
|
- re-visit lateral/thinning repartition of melting (modifs from May 04, 2007): |
| 502 |
|
prevent lateral melting if sea-ice fraction is too close to iceMaskMin ; |
| 503 |
|
and update experiment global_ocean.cs32x15.icedyn (+start from pickup files). |
| 504 |
|
o external_forcing_surf : extend the use of salt_EvPrRn & temp_EvPrRn (fresh-water |
| 505 |
|
tracer content) to linFS or no RealFreshWaterFlux ; default remains unchanged. |
| 506 |
|
o define mass2rUnit & rUnit2mass as units conversion factor, used for surface |
| 507 |
|
forcing. mass2rUnit: from mass per unit area [kg/m2] to r-coordinate unit |
| 508 |
|
(z-coord: =1/rhoConst ; p-coord: =gravity) ; rUnit2mass is just the reverse; |
| 509 |
|
and remove horiVertRatio (& recip_horiVertRatio). Affects truncation error |
| 510 |
|
of exp. global_oce_in_p: 10 16 16 14 16 13 16 16 16 13 13 13 16 13 13 12 16 |
| 511 |
|
-> update output.txt |
| 512 |
|
o pkg/cd_code: add brackets in multiple terms sum (S/R CD_CODE_SCHEME): |
| 513 |
|
this affects truncation error: -> update output.txt of exp: |
| 514 |
|
global_ocean.90x40x15 (fw&adm), lab_sea (fw&adm) & lab_sea.lsr, |
| 515 |
|
ideal_2D_oce, isomip.htd, natl_box, tutorial_global_oce_latlon. |
| 516 |
|
|
| 517 |
|
checkpoint59f |
| 518 |
|
o pkg/autodiff: zero_adj resets an ajoint variable to zero (in adjoint mode) |
| 519 |
|
o pkg/ctrl: ctrl_bound forces parameter corrections to stay in |
| 520 |
|
a bounded range (in forward mode) |
| 521 |
|
o pkg/ecco, ctrl, autodiff and mom_common: |
| 522 |
|
internal parameter controls, completements of implementations |
| 523 |
|
o add few EXCH S/R (UV_BGRID, UV_DGRID, SM) for B & D-Grid velocity & 2nd Moment Sxy. |
| 524 |
|
o use UV_BGRID exch for dxV,dyU: for now, only with pkg/exch2 & forward ; |
| 525 |
|
(not seem any difference in testing results). |
| 526 |
|
o pkg/cd_code: use the right EXCH calls (for CS-grid); but problem remains |
| 527 |
|
at face corners. |
| 528 |
|
o pkg/generic_advdiff: |
| 529 |
|
- in GAD.h, define as parameters (iMinAdvR,iMaxAdvR,jMinAdvR,jMaxAdvR) |
| 530 |
|
loop range indices for computing vertical advection tendency |
| 531 |
|
and do the calculation in the interior only (no overlap) ; |
| 532 |
|
for now, only implemented for SOM advection (to save CPU time). |
| 533 |
|
- add argument "withSigns" to S/R FILL_CS_CORNER_TR_RL (needed for SOM_xy moment) |
| 534 |
|
o exch2 adjoint |
| 535 |
|
First set of changes: |
| 536 |
|
- add exch2 hand-written adjoint templates |
| 537 |
|
- exch2 hand-written no longer use TAF-flag 'nownew_arg' |
| 538 |
|
This will require cleanup in pkg/autodiff/*.F |
| 539 |
|
(changed routines currently kept separate in verif. code/) |
| 540 |
|
- Comment all relevant #ifndef ALLOW_AUTODIFF_TAMC |
| 541 |
|
that used to hide exch2 or cubed-sphere specific code |
| 542 |
|
(commented via 'cph-exch2') |
| 543 |
|
o pkg/seaice |
| 544 |
|
- Close southern and western boundaries for UICE/VICE when useOBCS=.TRUE. |
| 545 |
|
o pkg/bulk_force: fix loading into Qsw of Short-Wave flux; and update |
| 546 |
|
output of global_ocean.cs32x15.thsice . |
| 547 |
|
|
| 548 |
|
checkpoint59e |
| 549 |
|
o exch2: new S/R that only use the cube-tracer (center-position) exchange: |
| 550 |
|
for B-grid vector (not yet tested) and C-grid vector (<- only used |
| 551 |
|
when option W2_USE_R1_ONLY in W2_OPTIONS.h is defined). |
| 552 |
|
o starting implementation of Duffy et al. (GRL 1999) salt plume scheme |
| 553 |
|
(#ifdef ALLOW_SALT_PLUME) |
| 554 |
|
o gather_2d/scatter_2d: removed halo regions |
| 555 |
|
o gmredi: add option (GM_taper_scheme='fm07') for Ferrari & McWilliams 2007 |
| 556 |
|
scheme and the corresponding test case (under front_relax dir). For now, |
| 557 |
|
only available with Skew-flux form and not yet tested in realistic set-up. |
| 558 |
|
adjoint: store directives updated, but gradient checks deteriorated, |
| 559 |
|
so maybe GM/Redi adjoint is now broken. |
| 560 |
|
o seaice adjoint |
| 561 |
|
* add AREA, HEFF, HSNOW as control variable |
| 562 |
|
(should not change lab_sea results) |
| 563 |
|
* CTRL_OPTIONS.h needs new flag: CTRL_SET_OLD_MAXCVARS_40 |
| 564 |
|
o pkg/seaice: |
| 565 |
|
* added computation of saltFlux in seaice_growth; |
| 566 |
|
For time being, sea ice salinity is assumed constant |
| 567 |
|
contribution of snow flooding to freshening of sea ice is neglected. |
| 568 |
|
SEAICE_salinity is set in SEAICE_PARM01 of data.seaice and it |
| 569 |
|
defaults to 0.0 in order to maintain backward compatibility. |
| 570 |
|
Typical values for SEAICE_salinity are 4 to 10 g/kg. |
| 571 |
|
* replaced dRf(1) with dRf(kSurface)*hFacC(kSurface) in seaice_growth.F |
| 572 |
|
* added SEAICE_availHeatFrac, which specifies fraction of surface level |
| 573 |
|
heat content used to melt or grow ice |
| 574 |
|
* commented out cap of HICE: it helps keep ice thinner |
| 575 |
|
* reintroduced cap of HEFF via MAX_HEFF and SEAICE_CAP_HEFF option |
| 576 |
|
* applied OBCS T/S mask to HEFFM sea ice mask |
| 577 |
|
|
| 578 |
|
checkpoint59d |
| 579 |
|
o standard name: {PKG}_DIAGNOSTICS_STATE to fill diagnostics array |
| 580 |
|
for {PKG} state variables, and called from DO_STATEVARS_DIAGS. |
| 581 |
|
(keep {PKG}_DIAGNOSTICS_FILL for non-state variables diagnostics) |
| 582 |
|
o pkg/gmredi (gmredi_readparms.F): add some missing "_d 0" |
| 583 |
|
and change default value of parameter "GM_Small_Number"; |
| 584 |
|
update output of exp: lab_sea (x2), cfc_example, global_ocean.90x40x15, |
| 585 |
|
global_ocean.cs32x15 (x4), global_ocean_ebm, global_with_exf, |
| 586 |
|
tutorial_global_oce_biogeo, tutorial_global_oce_latlon |
| 587 |
|
o pkg/ecco: cost function when applying pkg/smooth to controls |
| 588 |
|
o pkg/ctrl: add forcingPrecond run time parameter |
| 589 |
|
o pkg/seaice: |
| 590 |
|
- replace ZMIN=4.e8 with ZMIN=0. in dynsolver.F as in C-grid code |
| 591 |
|
- fix bug in metric terms of u-equation in LSR solver (implicit part), |
| 592 |
|
changes verification/lab_sea |
| 593 |
|
- rearrange implicit part of u-equation in LSR solver to make it easier to |
| 594 |
|
read, understand, and debug. This causes changes at the truncation level. |
| 595 |
|
- fix (serious) bug in the boundary conditions (sign error in the implicit |
| 596 |
|
part) of the LSR solver. Changes all LSR results. |
| 597 |
|
- fixed problem with HSNOW initialization to avoid thick snow when AREA |
| 598 |
|
is initilized with small values - does not change lab_sea results. |
| 599 |
|
- added HsnowFile initialization capability |
| 600 |
|
o pkg/smooth: addition to model/src and pkg/ctrl so allow |
| 601 |
|
initialization of pkg/smooth and application to control vector |
| 602 |
|
o pkg/seaice: fix another bug in the LSR no-slip boundary conditons |
| 603 |
|
o dependencies: pkg/profiles requires pkg/cal |
| 604 |
|
o pkg/profiles: |
| 605 |
|
- added option ALLOW_PROFILES_GENERICGRID |
| 606 |
|
-> pkg/profiles can thus be used with any grid (e.g. cs32x15) assuming |
| 607 |
|
that interpolation information (grid points and coefficients) is provided |
| 608 |
|
within the netcdf input files. |
| 609 |
|
-> sample matlab scripts can be found in MITgcm_contrib/gael |
| 610 |
|
that allow you to prepare netcdf input files. |
| 611 |
|
- also modified: cleaner include statements, cleaner error print statements, |
| 612 |
|
cleaner stops when error, and more tests of the netcdf input files consistency. |
| 613 |
|
o additional test (in global_ocean.cs32x15) using seaice+thsice pkgs. |
| 614 |
o pkg/seaice: |
o pkg/seaice: |
| 615 |
|
- move seaice-diagnostics from seaice_do_diags to do_statevars_diags; this |
| 616 |
|
requires a new routine: seaice_diagnostics_fill.F |
| 617 |
- make sure that seaice does not move if not forced (pressure replacement |
- make sure that seaice does not move if not forced (pressure replacement |
| 618 |
method, Hibler and Ip, 1995), changes results |
method, Hibler and Ip, 1995), changes results |
| 619 |
- and new damping criterion for EVP solver according to Hunke, JCP, 2001 |
- and new damping criterion for EVP solver according to Hunke, JCP, 2001 |
| 743 |
ATMOSPHERIC_LOADING is #define). |
ATMOSPHERIC_LOADING is #define). |
| 744 |
o pkg/kpp: replaced _KPP_RL with _RL |
o pkg/kpp: replaced _KPP_RL with _RL |
| 745 |
o bug fixes for vertical diffusivity computations when both KPP and |
o bug fixes for vertical diffusivity computations when both KPP and |
| 746 |
3D diffusivity arrays are used. |
3D diffusivity arrays are used; pkg/kpp/kpp_calc_diff_s & _t.F |
| 747 |
|
are no longer updating the vertical diffusivity array |
| 748 |
|
(passed as argument) but are (re-) setting those arrays. |
| 749 |
o pkg/seaice: fix bug in evp solver, while doing that |
o pkg/seaice: fix bug in evp solver, while doing that |
| 750 |
- change (improve?) discretization of stress computations for both |
- change (improve?) discretization of stress computations for both |
| 751 |
cgrid-lsr and evp solver |
cgrid-lsr and evp solver |
Parent Directory
| 
Revision Log
| 
Revision Graph
| 
Patch