--- MITgcm/doc/tag-index 2012/06/25 19:25:41 1.1675 +++ MITgcm/doc/tag-index 2012/08/17 21:30:51 1.1718 @@ -1,9 +1,272 @@ -$Header: /home/ubuntu/mnt/e9_copy/MITgcm/doc/tag-index,v 1.1675 2012/06/25 19:25:41 heimbach Exp $ +$Header: /home/ubuntu/mnt/e9_copy/MITgcm/doc/tag-index,v 1.1718 2012/08/17 21:30:51 jmc Exp $ $Name: $ Notes on tags used in MITgcmUV ============================== +o verification/testreport: + - add option '-tlm' for Tangent-Linear test (-adm,-oad,-tlm are all exclusive); + - can also use compressed (gzip) reference output file; + - change format (in pkg/grdchk) of TLM output (+ add forward gradient) + to be easier to collect by testreport. + - update existing TLM output (half out-of-date by more than 3 yrs). + - report NETCDF & LAPACK settings (from genmake2) in output dir. + - report run-time in each individual summary (previously only for AD test); + for TLM test, add also TAF version. + - add daily TLM test on (old) aces cluster + weekly test on baudelaire. + +checkpoint63r (2012/08/15) +o tools/genmake2: + - fix Makefile for target ftlall (+ update files to clean after make ftlall) +o pkg/ecco/ecco_cost_final.F : + - fix fc print statements (to stdout and costfunction0000) so + that the printed value is always after the global sum. +o pkg/seaice/seaice_dynsolver.F : + - fix ice strength assymptote at SEAICE_area_max<1 + (no change for the standard case of SEAICE_area_max=1) +o some cleaning: + - pkg source files: ensure that all *.F files from each pkg included the + corresponding ${PKG}_OPTIONS.h file (except stand-alone src file which + only includes CPP_EEOPTIONS.h); fix S/R CTRL_SMOOTH ; update AD experiment + code_ad/*.F src files; add new option-file in pkg/admtlm + corresponding + includes; + - model/src: add PACKAGES_CONFIG.h wherever ALLOW_AUTODIFF[_TAMC] is used. + - ALLOW_DIFFKR_CONTROL: add a stop in S/R CTRL_CHECK in case ALLOW_3D_DIFFKR + is undef; also define /DYNVARS_DIFFKR/ common block in DYNVARS.h when + ALLOW_DIFFKR_CONTROL is defined, to allow to compile and reach this stop. + - update option-file CPP_OPTIONS.h of all AD experiments: + * no longer include PACKAGES_CONFIG.h ; + * includes ECCO_CPPOPTIONS.h if PACKAGES_CONFIG.h has been included ; + * def/undef ALLOW_3D_DIFFKR independently of ALLOW_DIFFKR_CONTROL (before + including ECCO_CPPOPTIONS.h). +o pkg/cost, pkg/ctrl, pkg/ecco option-file: + - clean-up and split content of old pkg/ecco/ECCO_CPPOPTIONS.h file into + separated pkg option-file (done according to old doc from ECCO_CPPOPTIONS.h, + not necessary accurate). + - add a template for option-file ECCO_CPPOPTIONS.h, mostly empty, in + pkg/autodiff, i.e., the 1rst pkg to compile for adjoint built, with lines to + deliberately break compilation (a remainder that it needs to be customized) + - in pkg/ecco, includes ECCO_OPTIONS.h instead of COST_OPTIONS.h + - rename COST_CPPOPTIONS.h to COST_OPTIONS.h + - rename CTRL_CPPOPTIONS.h to CTRL_OPTIONS.h +o tools/genamke2: + - check for lapack. +o pkg/cost and pkg/ecco : attempt to sort out some of the limiting intricacies. + - pkg/cost.cost_final.F : allow for use of pkg/ecco and pkg/cost cost functions + jointly. The exclusive ifdefs were removed, to add costs terms (when defined) + from both packages. + - the fc global sum in pkg/ecco/ecco_cost_final.F was removed accordingly. + - for objf_atl we use either the pkg/ecco or the pkg/cost term depending on CPPs. + - pkg/ecco and model/src : add CPP brackets around pkg/cost and pkg/ctrl includes + of "CTRL_SIZE.h" "ctrl.h" "ctrl_dummy.h" "optim.h" / "cost.h" accordingly. +o pkg/cost : add comments to explain special and/or untested setups. +o pkg/exf: + - use EXF_OPTIONS.h to set pkg/exf options except if done in ECCO_CPPOPTIONS.h + (previously, ALLOW_AUTODIFF_TAMC was controling this switch). +o verification/*/code_ad: + mimimal cleaning of AD experiments ECCO_CPPOPTIONS.h option-file: + - reccord when this file is included (define ECCO_CPPOPTIONS_H) + - remove the most obvious obsolete features (e.g., #define ALLOW_COST not + allowed since Oct 2003 in any option-file except in PACKAGES_CONFIG.h) +o pkg/seaice : + - added possibility to change ice strength/PRESS0 relationship to + HEFF, according to the following run time parameters + SEAICEpresPow0 :: HEFF exponent for ice strength below SEAICEpresH0 (integer, >0) + SEAICEpresPow1 :: HEFF exponent for ice strength above SEAICEpresH0 (integer, >0) + SEAICEpresH0 :: HEFF threshold for ice strength (m) +o pkg/ggl90 : + - added run time flag mxlSurfFlag to include the code that ensure + mixing between first and second level (previously included as a comment) +o verification/1D_ocean_ice_column/input_ad/data.autodiff : + - add test of newly added dumpAdByRec=T option where + we write all records to one file per variable. +o verification/global_ocean.cs32x15/code_ad/CPP_OPTIONS.h : + - switch to #define ALLOW_BALANCE_RELAX and ALLOW_BALANCE_FLUXES + to check that this code is ok with taf and compiles in adjoint. +o pkg/autodiff : + - added run time param dumpAdByRec : ad dump format switch. + If true then write all records to one file per variable. + Else write one file per record (remains the default). + - revise seaice_ad_dump.F and exf_adjoint_snapshots_ad.F accordingly. +o pkg/seaice: + - Added SEAICEuseTILT run time switch : if set to + false then omit the tilt term in seaice dynamics. + - Fix free drift calculation. +o pkg/shelfice: + When initializing SHELFICEDragQuadratic, set default to be + consistent with shiCdrag in case SHELFICEuseGammaFrict is used, + but retain option to set it independently + (inconsistenty spotted by Veronique Dansereau) +o eesupp/inc: + - in EEPARAMS.h, add real-type constant for some frequently used simple number + (0,1,2,1/2), with name explicitly referring to RS/RL type. +o model/src: + - move balancing of relaxation terms to new S/R balance_relax.F. +o pkg/autodiff: + re-write active I/O high-level (control) S/R: + - rename S/R ACTIVE_READ/WRITE_RL to ACTIVE_READ/WRITE_3D_RL + - add 1 more argument (useCurrentDir) to replace the "_LOC" 2nd set of S/R. + - use standard pkg/mdsio I/O routines (MDS_READ/WRITE_SEC_XZ/YZ and + MDS_READ/WRITE_FIELD); change loops to multi-threads. + - update and merge active_file_control_slice.flow into active_file_control.flow + - update the calls in lower level S/R (active_file.F, active_file_g.F and + active_file_ad.F) keeping the argument list and function unchanged. + - also update lower-level "_LOC" version (call same S/R with "useCurrentDir + =.TRUE.") and remove high-level (control) _loc version (not called anymore). + - new low-level active I/O routines, one for each type (RL/RS), for active var + of various (XY,XZ,YZ) shape and level number (arguments: arrShape, myNr); + fix ctrl_init_wet.F & ecco_cost_weights.F by calling the new S/R with the + right type. +o autodiff and RS=real*4: + - call the right type (_RS instead of _RL) of S/R in few output S/R (pkg/exf + ad-monitor and snapshots); disable arrType='RS' in mdsio_gl*.F ; + - change argument type of 3-D mask argument to _RS to match type in the call + (in seaice_cost*.F, ctrl_*_gen.F and *ctrl_bound.F); to be continued. +o pkg/thsice: + - re-anable balanceEmPmR when useSEAICE=T with useThSIce=T. + - option to correct ocean fresh-water flux for imbalance of Atmos+Land + fresh-water flux: uniform correction or proportional to local precip. +o pkg/ctrl: + - new header file CTRL_SIZE.h for parameters maxCtrlArr2D, maxCtrlArr3D + - redefine CPP option ALLOW_BOTTOMDRAG_CONTROL to + ALLOW_BOTTOMDRAG_CONTROL_NONGENERIC + - verification/hs94.1x64x5/ tests ALLOW_GENARR{2D,3D}_CONTROL + +checkpoint63q (2012/07/30) +o pkg/autodiff: + note about monitor and snap-shot of ad-variables: + a) need to apply ADEXCH before monitor or output of ad-vars to include + overlap contributions (was done for adjDump but missing for monitor). + b) since ADEXCH implies adding overlap values to interior array, machine + truncation error is expected when adding an ADEXCH call in the middle of + the time-stepping (although mathematically identical). + modifs: + - fix some incorrect type (RL/RS, UV/scalar) in ADEXCH calls and ad-vars + declaration (adEddyPsiX,adEddyPsiY in adcommon.h). + - add new run-time params (dumpAdVarExch & mon_AdVarExch) to select how to + apply ADEXCH before monitor or snap-shot of ad-variables: + =0 : no ADEXCH ; =1 : do ADEXCH ; =2 : apply ADEXCH on a local copy. + - implement new method (copy+adexch) for snap-shot and monitor of main model + ad-vars (initially not used by default -> output unchanged). + - change default (set to 2) to use new method and update few output_adm.txt + with accurate ad-monitor (global_ocean.90x40x15, tutorial_global_oce_biogeo + and both tutorial_tracer_adjsens tests). + - switch 2 global_ocean.cs32x15 AD-tests to AB-2 (seaice_dynmix & thsice) + and turn off momDissip_In_AB in the 2 other (needed for stable AB-3). +o pkg/ctrl: + enable the use of a variable number (maxCtrlArr2D, maxCtrlArr3D) of + generic control variables via + #define ALLOW_GENARR2D_CONTROL and/or #define ALLOW_GENARR3D_CONTROL + Only required customization should be S/R ctrl_map_ini_genarr.F + (routine in repository contains example) +o pkg/exf: + - fix and refine checking of surface-relaxation setting when using EXF +o verification (adjoint exp): + - reduce length of simulation by half for experiment global_ocean.cs32x15 + (std, seaice & seaice_dynmix) and exp tutorial_dic_adjoffline; update output. +o pkg/mom_common (mom_u/v_bottomdrag), pkg/shelfice (shelfice_u/v_drag) + - replace if statements KE.ne.0 with KE.gt.0 to avoid sqrt of negative numbers + of KE that can occur with aggressive optimization + + put quadratic bottom drag calculation in separated loops (only done if + using bottomDragQuadratic) +o model/src: CG2D_NSA + - clean-up: stop if cg2dNormaliseRHS=T and AUTODIFF + and replace GLOBAL_SUM calls with GLOBAL_SUM_TILE. + - in test-exp that uses CG2D_NSA (bottom_ctrl & tutorial_tracer_adjsens), + change cg2d residual target to be expressed in Wunits (=> do not normalise + RHS ; this fix global_max Pb with TAF); update output. +o verification/tutorial_tracer_adjsens: + improve parameter consistency (and get rid of warnings): + - use deltaTFreeSurf=deltaTtracer (better with NonLin FreeSurf); + - switch Temp Adv Scheme from 81 to 80 (no limiter since theta < 0 is physical) + - use staggerTimeStep (for stability) in som81; use default rotation period. +o pkg/exch2: + - with empty facet: fix compact-format (W2_mapIO=1) definition (was previously + reset to 0 = 1 long line in X) +o pkg/mdsio: + - fix reading of dimensions from meta file for large-size domain +o model/src: + - fix saltIsActiveTr setting for IdealGas eosType (eosType is 6 character long); + - to refine CS-grid check, add logical flag "hasWetCSCorners": true if using + Cubed-Sphere Exch with CS-corners inside the domain; + make new S/R INI_GLOBAL_DOMAIN from code in ini_linear_phisurf.F to calculate + globalArea and to set hasWetCSCorners. + - add warning if usingCurvilinearGrid and momAdvection in flux-form (missing + metric terms); stop if hasWetCSCorners=T and momAdvection in flux-form. + - re-activate the stop if useCDscheme and hasWetCSCorners=T (previously if + useCDscheme and useCubedSphereExchange). +o pkg/seaice/seaice_growth.F : + - compute the SEAICE_DISABLE_HEATCONSFIX term accordingly + for the (.NOT.useRealFreshWaterFlux).OR.(nonlinFreeSurf.EQ.0) + when the previous code was not correct. + - in all cases diagnoze the boundary condition mismatch to SIaaflux + - in the case of useRealFreshWaterFlux.AND.(nonlinFreeSurf.GT.0) add + that term to QNET (essentially to set the melt/freeze water to 0degC). + when SEAICEheatConsFix is true, so that the ocean+ice budget heat is closed. +o verification/global_ocean.cs32x15/input.seaice : + - add SEAICEheatConsFix in data.seaice, ensuring closed ocean+ice heat budget. + - add diagnostics (commented out) for budget test using my matlab codes (gcmfaces). + - update results (output.seaice.txt) accordingly. +o verification/lab_sea/input/data.seaice : + - remove SEAICEheatConsFix, which is now be tested + in global_ocean.cs32x15. The presently coded fix + is only consistent with real fresh water + nlfs. + - update results (output.txt) accordingly. +o pkg/salt_plume: added diagnotic for salt tendency (oceSPtnd). +o AD_CONFIG.h file: + - in genmake2: remove "#undef ALLOW_ECCO_OPTIMIZATION" from AD_CONFIG.h; + - remove #include "AD_CONFIG.h" from all included OPTIONS files and + add it explicitly in each fortran src file where it is needed. + motivation: header file AD_CONFIG.h is generated/modified after "make depend" + thus introducing potentially wrong dependency; therefore, better to include + it only where it's really needed, and after any other included OPTIONS file + (less tempting to bring conditional includes according to ALLOW_ADJOINT_RUN + or ALLOW_TANGENTLINEAR_RUN). + +checkpoint63p (2012/07/06) +o pkg/autodiff: + - rename S/R INI_AUTODIFF to AUTODIFF_INIT_VARIA ; + - create new header file "AUTODIFF_PARAMS.h" for pkg/autodiff parameters + which are now read from new parameter file "data.autodiff" (inAdExact + previously in "data", use{KPP/GMRedi/SEAICE}inAdMode previously in data.pkg) + - add file "data.autodiff" in all verification/*/input_ad/ directories. + - add run-time parameter to turn On/Off SEAICEuseFREEDRIFT in adjoint/forward + mode (used in global_ocean.cs32x15.seaice_dynmix test exp.) +o verification/tutorial_tracer_adjsens/ + - add extra sub-verification which uses "SOM" advection scheme + (2nd Order-Moment Advection Scheme, Prather Limiter) with + #define GAD_ALLOW_TS_SOM_ADV in GAD_OPTIONS.h + and in data: + multiDimAdvection=.TRUE., + tempAdvScheme=81, + saltAdvScheme=81, +o model/src: + - introduce ALLOW_BALANCE_RELAX which allow the removal + of the global mean of relaxation terms by setting + balanceThetaClimRelax and balanceSaltClimRelax + - disable balanceEmPmR and balanceQnet in the case when useSeaice. + This case is now treated appropriately in seaice_growth.F +o pkg/seaice: + - do balanceEmPmR and balanceQnet based on the net fluxes + (SItflux) that enter the ocean+ice system rather than QNET + and EMPMR. Works best in real fresh water. + - SIatmFW, SIatmQnet, and SItflux computations had to be moved + out of the diagnotics bloc, and stored in global arrays. + +checkpoint63o (2012/06/29) +o vertical mixing schemes: + - remove unused S/R CALC_BUOYANCY (left from branch-atmos merging) + - replace computation of density (in Function STATE1, opps_calc.F) by a call + to S/R FIND_RHO_SCALAR + - pass sigmaR to S/R PP81_CALC, MY82_CALC (but not yet used) and to GGL90_CALC + and use it for N^2 calculation in ggl90_calc.F (save 2 density computations); + affects output (vermix.ggl90) at truncation level. +o testreport: + - change adjustment of SIZE.h for mpi+mth testing: preserve enough tiles + (nSx,nSy) for the number of threads (from eedata.mth) +o model/src: + - remove 2nd set of calls at the top of initialise_varia.F (Adjoint part) + related to Non_Lin Free-Surf or Bottom-Control (but not longer necessary). o verification/tutorial_tracer_adjsens/ - compile with ALLOW_CG2D_NSA - run with nonlinFreeSurf=3