--- MITgcm/doc/tag-index 2009/06/01 14:22:57 1.1114 +++ MITgcm/doc/tag-index 2009/11/10 09:34:04 1.1191 @@ -1,9 +1,270 @@ -$Header: /home/ubuntu/mnt/e9_copy/MITgcm/doc/tag-index,v 1.1114 2009/06/01 14:22:57 jmc Exp $ +$Header: /home/ubuntu/mnt/e9_copy/MITgcm/doc/tag-index,v 1.1191 2009/11/10 09:34:04 mlosch Exp $ $Name: $ Notes on tags used in MITgcmUV ============================== +o pkg/seaice + - fix diagnostics of principal stress components for EVP + - add optional diagnostics of non-linear convergence for LSOR solver + - make maximum zeta a runtime parameter +o model/src (3-d solver): + Undo modif from Dec 12, 2005 (solve_for_pressure.F cvs version v.49), + which consisted in removing the Surf. Non-Hydrostatic pressure from the + whole column and adding it to Eta (inspired from Casualli, 1999): + Only affects 3-D solver initial guess and only if using ExactConserv, + but tests did show a degradation of solver convergence. For this reason, + decide to disable this piece of code. + to recover previous version, just set zeroPsNH = exactConserv (line 78). + Update output of exp. global_ocean.cs32x15.viscA4 & hs94.cs-32x32x5.impIGW +o pkg/ggl90: + - fix previous modif (unfortunatly has not been documented here). +o pkg/seaice/autodiff: + - add CPP flag to include clipping of zeta in EVP code; + although clipping of zeta is not necessary for EVP and not recommended + it was done by default, now clipping is disabled by default + - modify computation and store zetaC/Z in seaice_evp for TAF + - add CPP-flag SEAICE_ALLOW_CLIPVELS (undefined by default) + - change seaice_calc_viscosities/strainrates for better + vectorization after AD by TAF + - rearrange common blocks in SEAICE.h once again for easier dumps of + adjoint output, adjust adcommon.h and g_common.h + - change flux form of default advection scheme for better vectorization + of adjoint code + - modify lab_sea/input.lsr to test the flux form of default advection + scheme +o autodiff: fix seaice fields in g_common.h (left over from 3 to 1 level + transition in checkpoint61r) + +checkpoint61x +o adjoint: adding two new verification experiments + * tutorial_global_biogeo: an online dic sensitivity experiment + (work with G. Forget) + * tutorial_dic_adjoffline: an offline dic sensitivity experiment + (an update of the Dutkiewicz et al., 2006 paper) + Both use the same CO2FLUX cost function, but different controls + Gradient checks look good, but some recomp. could still be traightened out + Should add xx_dic control variable to test at some point +o model/src: + add run-time parameters to check and/or mask initial Temp & Salt +o pkg/ptracers: + apply zonal filter (if used) to all passive tracers + +checkpoint61w +o adjoint: + * update and finish work on adjoint of dic package (G. Forget) + * start adjoint of offline package + * bug fix for cube-sphere TLM code (active_file_g.F) +o tutorial_global_oce_optim: + now tested with 4 tiles (change data.grdchk and update output_adm.txt) + and add SIZE.h_mpi for MPI testing +o pkg autodiff + - fix bug in autodiff_restore + - make genmake2 remove comments and empty lines from ad_input_code.f in + order to reduce file size (new script file remove_comments_sed) +o allows to specify vertical profile of vertical viscosity (viscArNr) +o pkg autodiff + autodiff_ini_model_io: comment out writing of grid information to netcdf + files, because at this time the grid fields are a bunch of zeros + (this is analogous to ini_mnc_vars.F) +o pkg seaice/autodiff: + left-over from removing the 3-time-level fields: + - small rearrangement of common blocks (SEAICE.h) + - fix field declaration in order to get meaningful output in + adseaice.*.nc (adcommon.h) +o pkg obcs: + - go through NH code only if nonHydrostatic=T + - fix restart for Orlanski (Hydrostatic; still to fix NH+Orlanski restart) + - move call to OBCS_CALC before SEAICE_MODEL which needs seaice-obc fields + => Update output of experiment seaice_obcs + - useOBCSbalance: code is wrong with multi-tile / proc: add a stop. + - fix NH+Orlanski restart (add wvel OBC value in pickup file) + - fix restart for useOBCSprescribe + useEXF +o pkg ctrl + - make obcs as control parameter work also with useSingleCPUio + - replace a few sny and snx by Ny and Nx to be consistent with + ctrl_set_globfld_x/yz.F +o pkg seaice + - re-introduce global field frWtrAtm for the case of + ALLOW_MEAN_SFLUX_COST_CONTRIBUTION defined (SEAICE.h, seaice_growth.F) + - catch the case of an empty filename in seaice_cost_init_fixed.F +o pkg ecco: + - small bug fixes in the_main_loop + . fix order of monitor arguments + . change SEAICE_MULTILEVEL to SEAICE_MULTICATEGORY + . move include DYNVARS.h up outside the ALLOW_AUTODIFF_TAMC block so + that ALLOW_ECCO_FORWARD_RUN works with NONLIN_FRSURF + - replace cos(yc*deg2rad) by _rA in cost_mean_heat/saltflux.F, fix the + imbalance diagnostics for wmean_h/sflux = 0 + - add actual values to f_obcsn/s/w/e in ecco_cost_final.F + - catch the case of empty filenames in ecco_cost_init_fixed.F + and ecco_readparms.F + - fix the logic of wsalt=0.and.wsalt2=0 to wsalt=0.or.wsalt2=0 (and theta) + to make this statement meaningfull in ecco_cost_weights.F +o pkg zonal_filt: + - change ZONAL_FILTER S/R interface (allows to filter 2-D fields) + - when using rStar, filter uVel*hFacW instead of uVel + (tested without topography: fix a Pb of growing instability with NLFS) +o pkg/aim/thsice: + -Add capacity to read a Q-Flux to be applied in slab-mixed layer thsice_slab_ocean.F + (works as SST restoring) +o pkg/thsice: + -add specific parameter for salt restoring in slab mixed-layer + (by default equal to the temperature restoring) + +checkpoint61v +o pkg/kpp: + -add double diffusive contributions as a hack, turned off by default + for now, and the code can be excluded with a CPP-flag + EXCLUDE_KPP_DOUBLEDIFF just as EXCLUDE_KPP_SHEAR_MIX + -add a verification experiments to "vermix" to test this code +o pkg/mom_common: + Move computation of length scales to mom_init_fixed.F (avoid recomputation + at each time-steps*levels). +o pkg/layers: + Created a new package for computing volume fluxes in temperature or + [not yet implemented] density layers, e.g. for calculating residual + overturning circulations. +o pkg/diagnostics: + implement a RS type version for the set of DIAGNOSTICS-FILL subroutines +o fix few RS/RL type mismatch (check with ifort 11.1 "-warn all" option) +o pkg/mdsio (& pkg/rw): rework high level S/R interface: + To avoid mixing type (RS/RL) of input/output array argument, + replace single mixed-type array with a pair from each type (RS/RL). +o setting pickupSuff force to start from a pickup (even if nIter0=0): this + is now implemented for all pkgs (previously only done for the main pickup) + +checkpoint61u +o pkg/obcs: + - change OB?eta from _RS to _RL and put them into a separate commen block + - add code to read OB?eta from a file via obcs_prescribe_read: the code + compiles and does not destroy any other experiments, the data is read + properly, but the code is not tested in actual applications +o ini_mnc_vars.F: if available (ALLOW_CAL), use startTime_1 and 2 to + create a meaningful unit attribute for the time coordinate variable +o pkg/ebm: add plenty of missing "_d 0" ; + update results of experiment global_ocean_ebm (fwd + adjoint tests) + Notes: several pieces of code only works for (very) specific set up + (domain size, lat range, etc ...); should be documented & stop if one + of those assumptions is not meet. +o pkg/seaice: add missing "_d 0" in HSNOW initialisation + update results of experiments: lab_sea (3 fwd tests + 2 adjoint tests) + and offline_exf_seaice (seaicetd + adjoint). +o add missing "_d 0" in ShortWave flux penetration (S/R SWFRAC): + update results of experiments: lab_sea (4 fwd tests + 3 adjoint), + seaice_obcs, global_ocean.cs32x15 (icedyn & thsice) and natl_box (x2). +o few changes to pass when compiling with strick checking of arguments + across S/R (allow to find few bugs in S/R calls). +o pkg/seaice: + - seaice_evp: move if-statement out of loop (TEM) at the cost of 11 new + 2d-fields, for slightly better performance + +checkpoint61t +o pkg/seaice: + - remove unused variables in seaice_evp + - reduce number of SQRT in seaice_evp and seaice_calc_viscosities +o eeset_parms.F, open_copy_data_file.F & ini_parms.F: + named file replacing SCRATCH files (for TARGET_BGL & TARGET_CRAYXT): + each processor opens a different file (a fix from Matt). +o cg2d: + use simple EXCH (overlap size = 1 and ignore corners), like in cg3d.F : + this reduces number of EXCH calls by 2 when using exch2. + For now, keep previous size array for gc2d_nsa.F to avoid Pb with store + directives and missing simple EXCH (exch_s3d_rx) adjoint code. +o pkg/thsice: + add code for Near Infra-Red albedo (from Jeff) + (turned on with run-time flag: thSIce_calc_albNIR, in data.ice). +o ini_model_io: + move (from packages_readparms.F & initialise_fixed.F) to ini_model_io.F + MNC & MONITOR initialisation calls; move MNC calls from {pkg}_readparms.F + to {pkg}_init_fixed.F for {pkg}= kpp, seaice & shelfice. + +checkpoint61s +o pkg/exch2 (exch2_rxN_cube*): + remove last BARRIER (no need to synchronise after getting data from shared + buffer (get) as long as any change to buffer (put,recv) is between BARRIER) +o change tiling of natl_box (from 2 to 4 tiles) to get it tested with MPI+MTH + and generate new output (for the 2 tests) using default rotationPeriod. +o pkg/exch2: + - always call exch2_*_cube, not exch-1 anymore, if useCubedSphereExchange=F + (was already the case with AUTODIFF_EXCH2 defined) + - add bj in exch2 arrays and S/R. + - can now use pkg/exch2 for regular (non CS) set-up; switch test experiment + MLAdjust to exch2 (identical results except with MPI -> truncation diff); +o add pkg/longstep for ptracer timestep a multiple of U/V/T/S timestep + - adds deltaT argument to the subroutines TIMESTEP_TRACER DWNSLP_APPLY + GAD_ADVECTION GAD_CALC_RHS GAD_IMPLICIT_R GAD_SOM_ADVECT + - verification experiment natl_box.longstep +o build_options: ifort+mpi_aces optfile (following Constantinos advise): + allow different versions of ifort (8 or 9) and different versions + of mpi (mpich or mpich2) to be used with the same optfile. + requires to load a netcdf module and set MPI_INC_DIR (see comments inside + optfile). Change testing scripts (example_scripts/ACESgrid) accordingly. +o pkg/seaice: + fix previous modif (3-level field removed) for old-pickup reading + +checkpoint61r +o pkg/seaice: + - clean up the 3-time levels of UICE,VICE,HEFF,AREA in three steps: + 1. introduce fields for time step (N-1) + 2. do not use levels 2 and 3 anymore (but keep them) + 3. replace 3D versions of UICE,VICE,HEFF,AREA by 2D versions. This + includes changes to pkg/thsice/thsice_get_velocity.F, + model/src/external_forcing_surf.F, pkg/obcs/obcs_apply_seaice.F, and + pkg/obcs/obcs_apply_uvice.F. + - completely remove code for leap-frog time stepping for second order + advection + - adjusted adjoint accordingly; + gradient checks have improved by an order of magnitude! + +checkpoint61q +o pkg/seaice: fix bug in free slip boundary conditons for LSOR solver + - the boundary conditions along open boundaries are unclear, because the + associated masks are set to zero and are thus treated as closed when + applying the slip-boundary conditions. + - move initialisation of grid variables from seaice_init_varia to _fixed +o adding doc/diags_changes.txt +o model/src: + do_the_model_io: rename few pkg S/R which are called from do_the_model_io, + to avoid "_diags" naming if no relation to pkg/diagnostics. + ini_theta,ini_salt: fix (at least compile the code) reading from NetCDF + file and only call PLOT_FIELD_XYZ if debugMode=T. + remove few PLOT_FIELD_XY calls outside initialisation. + remove few unnecessary BARRIER. +o pkg/salt_plume: add SaltPlumeSouthernOcean to allow disabling + of salt_plume package in Southern Ocean +o pkg/thsice: + - fix hOceMxL in overlap (needed for sea-ice advection); + update output of experiment global_ocean.cs32x15.icedyn. +o pkg/seaice: + - make the (placeholder-)tracer IceAge a little more meaningful + by avoiding ridging of age due to convergent ice flow + - add diagnostics for surface temperature, atmospheric and oceanic + heat flux, and atmospheric and oceanic ice growth + +checkpoint61p +o eesupp (global_max & global_sum): + - add one more element to buffer (start at index 0) for GL_MAX/SUM output; + - remove starting & endding barrier (no longer needed) +o pkg/mdsio, writelocal: + works also in multi-threaded when LOCBIN_IO_THREAD_SAFE is undefined + (remove the STOP): uses shared buffer IO to store data from all threads + and then let master-thread write nThreads tiles. +o pkg/mdsio (read/write field): + - tiled IO done in 1 piece (all levels at a time); + - new header file "MDSIO_BUFF_3D.h" with 3-D buffers. + - multi-threaded: allow to read/write local (non-shared) array + (was already working with singleCpuIO ; now works also without); + => no longer needs barrier call after reading a file (added inside + mds_read/write field). + - move barrier calls outside gather/scatter_2d to mds_read/write_field +o pkg/obcs/exf + - exf_getffieldrec.F: add a few ". _d 0" and + add code to catch the case fldperiod = 0. + - obcs_precribe_read.F: change if statements so that they match + the variable type (_RL) + - add a separate exf_iprec_obcs to exf_set_obcs (by default equal to + exf_iprec) as part of namelist EXF_NML_01 o pkg/mdsio: - read/write tiled (local) files: read/write 1-level tile chunk at a time (instead of segment of length sNx); expected to speed up tiled IO. @@ -18,7 +279,9 @@ - implement EXCH_IGNORE_CORNERS in scalar exchange (rx1); remove 2nd exch call in exch2_s3d_rx (no longer needed). - document S/R arg. list (+ update some comments) -o pkg/seaice: +o pkg/seaice: + - added SEAICE_CLIM_AIR code, which permits modulation of surface air + temperature and humidity over sea ice based on climatological values. - clean up computation of Hibler+Bryan (1987) stress coupling for the case of LSR (this change is expected to change the results slightly because now it uses slightly difference moduli from the second last LSR solution, but