/[MITgcm]/MITgcm/doc/tag-index
ViewVC logotype

Diff of /MITgcm/doc/tag-index

Parent Directory Parent Directory | Revision Log Revision Log | View Revision Graph Revision Graph | View Patch Patch

revision 1.1517 by jmc, Sat Oct 22 14:11:04 2011 UTC revision 1.1540 by jmc, Thu Dec 1 19:33:51 2011 UTC
# Line 4  $Name$ Line 4  $Name$
4      Notes on tags used in MITgcmUV      Notes on tags used in MITgcmUV
5      ==============================      ==============================
6    
7    o pkg/generic_advdiff (implicit_r):
8      - fix implicit vertical advection conservation with AB and/or NonLin-FreeSurf
9        (remove Tr*d/dz(w) in gad_implicit_r.F and add it in gad_calc_rhs.F);
10      - fix implicit vertical diffusion or advection with NonLin FreeSurf
11        (was using current recip_hFacC instead of future recip_hFacC): set future
12        recip_hFac in thermodynamics.F and pass & use it in all vertical implicit
13        solver S/R ;
14      - fill diagnostics of vertical advective fluxes (computed from tendency)
15        when using implicit vert. advection;
16      - update output of exp that uses NonLin-FreeSurf & implicit vertical diffusion:
17         global_ocean.90x40x15 (fwd+AD), global_ocean.cs32x15 (4 fwd +4 AD),
18         tutorial_global_oce_in_p (fwd), tutorial_tracer_adjsens (AD)
19         and cpl_aim+ocn (used in ocean); to be fixed: bottom_ctrl_5x5.
20    o pkg/diagnostics:
21      - add diags for T & S tendency which goes through Adams-Bashforth.
22    
23    checkpoint63f (2011/11/30)
24    o pkg/obcs/obcs_calc_stevens.F:
25      - join some loops to reduce number of if-statements
26      - fix a bug in code for northern and southern boundaries.
27    o pkg/seaice/seaice_init_varia.F
28      - undo resetting UVM=0 that has slipped in unnoticed in r1.49 and
29        caused zero ice-velocities for the B-grid code
30    o pkg/seaice/seaice_init_fixed.F
31      - fix a bug in the initialisation of the second metric terms
32        k2AtC and k2AtZ: k2AtC was set to the value of k2AtZ and k2AtZ was 0.
33      - this affects all seaice configurations with useCurviLinearGrid = .true.
34        and SEAICEuseDynamics = .true. (default) and SEAICEuseMetricTerms
35        = .true. (default)
36      - the only verification experiment affected is global_ocean.cs32x15:
37        icedyn, seaice, and seaice_dynmix (AD)
38    o pkg/thsice/thsice_diagnostics_state.F
39      - only for TARGET_NEC_SX: initialise all of tmpFld, so that overlap does
40        not contain garbage that can lead to surprises in diagstats_calc.F
41    o model/src:
42      - move bi,bj loops inside INTEGR_CONTINUITY ;
43      - call INTEGR_CONTINUITY directly from forward_step.F
44        (previously called from momentum_correction_step.F)
45      - call UPDATE_ETAH from integr_continuity.F and update initialise_varia.F,
46        forward_step.F and pkg/ecco/the_main_loop.F
47    o model/src/do_oceanic_phys.F:
48      - reset addMass to zero only if useICEFRONT ;
49      - separated k-loop for rhoInSitu calculation ;
50      - add 2 diagnostics for wVel*delta^k(Rho) at constant pressure (= WdRHO_P)
51        and at constant T,S (= WdRHOdP);
52      NOTE: change diagnostics WRHOMASS : this is now simply the Vertical transport
53        of In-Situ Density (Anomaly) (=wVel*rho^bar-k) (previously was Vertical
54        transport of locally-referenced potential density);
55        to recover old diag: (WRHOMASS)_old = (WRHOMASS)_new -1/2 * (WdRHOdP)
56    o gluemnc:
57      - replace ncrcat with ncks when adding variables to *.glob.nc
58      - change the way variables without record dimensions are handled so that
59        files that have variables with and without record variables can be glued:
60        add a record dimension with a new unique name to be able to differentiate
61        between variable with and without a-prior record dimensions. Remove
62        this record dimension before adding the variable to the *.glob.nc file.
63        This procedure automatically works also for grid.*.nc files that do not
64        have any record dimension.
65    o verification/cpl_aim+ocn:
66      - generate new reference output on baudelaire (gfortran, mpich2) (previously
67        from old aces-grid cluster, ao + ifort) and update 'run_cpl_test' script.
68    o pkg/obcs/obcs_calc_stevens.F:
69      - add code for northern and southern boundaries.
70      - reorder some loops to save some operations and so that drFBar can be scalar
71    o model/src:
72      - re-arrange extrapolation calculation in Adams-Bashforth S/R (affects machine
73        truncation) and add diagnostics for Adams-Bashforth tendency increment.
74      - update test-exp. AD output:
75        1D_ocean_ice_column, bottom_ctrl_5x5, lab_sea.noseaicedyn, obcs_ctrl;
76      and FWD output: exp4 (x3), fizhi-cs-32x32x40, fizhi-cs-aqualev20,
77        front_relax (x3), global_ocean.90x40x15 (x2), global_ocean.cs32x15 (x5),
78        global_ocean_ebm, ideal_2D_oce, internal_wave, isomip (x3), lab_sea (x4),
79        matrix_example, MLAdjust (x5), natl_box (x2), tutorial_advection_in_gyre,
80        tutorial_global_oce_in_p, tutorial_global_oce_latlon and
81        tutorial_plume_on_slope.
82    o pkg/diagnostics/diagstats_calc.F: add special code for TARGET_NEC_SX
83      that vectorises. This code uses 3 extra 2D fields, F90 array assignments
84      and F90 intrinsic routines SUM, MINVAL, MAXVAL, but since TARGET_NEC_SX
85      does not even come with an F77 compiler anymore, this should not be
86      a limitation
87    
88    checkpoint63e (2011/11/06)
89    o pkg/seaice adjoint of LSR:
90      - substantially rework STORE dirs.
91      - add new flag SEAICE_LSR_ADJOINT_ITER and parameter SOLV_MAX_FIXED
92        this flag will do SOLV_MAX_ITERS but only store SOLV_MAX_FIXED fields
93        and only perform SOLV_MAX_FIXED adjoint iters.
94        If flag SEAICE_LSR_ADJOINT_ITER is undef-ined, the store in the iteration
95        always goes to same record, meaning that the last forward field will be
96        the one used in the adjoint - an approximation.
97      - now produces non-zero LSR-related gradients, and acceptable gradients for
98        An's zonal (x-z) test case (to be checked in)
99      - lab_sea: produces acceptable gradients for lab_sea with forward/backward LSR
100        (but these are screwed up by the thermodynamics in the first place);
101        reduce from 4 to 3 timesteps to avoid t.d. issue as temporary solution;
102        output_adm.txt also reflects changes to seaice_lsr
103      - global_ocean.cs32x15:
104        * the "default" input_ad.seaice has been re-defined to always exclude dynamics
105          (output_adm.seaice.txt was updated)
106        * new input_ad.seaice_dynmix tests a forward LSR / backward free-drift
107          (output_adm.seaice_dynmix.txt)
108    o test-exp. exp4+stevens:
109      - switch off pTracers and nonHydrostatic (restart problems with NH); update output
110    o pkg/seaice (lsr solver):
111      - put coriolis term into FORCEX,Y (as in seaice_evp.F; previously in rhsU,V only);
112        update output (change truncation) of FWD test exp lab_sea & global_ocean.cs32x15
113      - fix SEAICEuseFREEDRIFT for new OBCS version (ndef OBCS_UVICE_OLD)
114      - new parameter "LSR_mixIniGuess" allows (if=2,4) to mix free-drift solution
115        into initial guess for linearised solver; use weight ~ (1/err)^2,4.
116      - compute (with new S/R) and print initial (+ free-driff sol. if available)
117        and final residual of linearised system.
118    o pkg/pp81: add diagnostics for vertical eddy viscosity and diffusivity
119      (requires new subroutines pp81_init_fixed.F/pp81_diagnostics_init.F)
120    o pkg/obcs: Stevens BCs (still only for east and west)
121      - fix Stevens BCs by using intermediate velocities of previous
122        time step as "baroclinic" contribution rather velocity of one grid cell
123        inside the boundaries; more consistent but makes the entire treatment
124        more complicated, requires new pickup files and two new S/Rs:
125        obcs_copy_uv_n and obcs_save_uv_n
126        keep old version with flag OBCS_STEVENS_USE_INTERIOR_VELOCITY
127        (undefined by default)
128      - silently move computation of some grid factors into separate loops to
129        save some recomputations
130      - add a hack for non-linear free surface
131      - add a verification experiment to exp4 that tests this code
132    o make gluemnc/xplodemnc work with (moron) non-GNU shell tools
133    
134    checkpoint63d (2011/10/25)
135  o pkg/obcs + seaice-dynamics:  o pkg/obcs + seaice-dynamics:
136    - new implementation calls OBCS_APPLY_UVICE (in seaice_dynsolver.F) before    - new implementation calls OBCS_APPLY_UVICE (in seaice_dynsolver.F) before
137      solving for uIce,vIce and prevent SEAICE_LSR or SEAICE_EVP to modify OB      solving for uIce,vIce and prevent SEAICE_LSR or SEAICE_EVP to modify OB
# Line 24  o pkg/layers: Line 152  o pkg/layers:
152  o pkg/dic:  o pkg/dic:
153    - Remove subroutine CALC_PCO2_APPROX_CO3 from carbon_chem.F and add    - Remove subroutine CALC_PCO2_APPROX_CO3 from carbon_chem.F and add
154      carbonate computation/output to CALC_PCO2_APPROX      carbonate computation/output to CALC_PCO2_APPROX
155    o pkg/generic_advdiff
156      - make TAF generate code that vectorizes:
157        gad_dst3fl_adv_*.F, gad_u3c4_impl_r.F
158  o model/src:  o model/src:
159    - improve vectorization also for adjoint code of    - improve vectorization also for adjoint code of
160      solve_tridiagonal and solve_pentadiagonal. As this requires up to      solve_tridiagonal and solve_pentadiagonal. As this requires up to
161      5 new 3D-fields, this code is only turned on for the combination of      5 new 3D-fields, this code is only turned on for the combination of
162      ALLOW_AUTODIFF_TAMC and TARGET_NEC_SX.      ALLOW_AUTODIFF_TAMC and TARGET_NEC_SX. Also remove error message
163        for singular matrix check (only for TARGET_NEC_SX).
164    - remove hard-wiring of CPP-flag ALLOW_SOLVERS_KLOOPINSIDE to    - remove hard-wiring of CPP-flag ALLOW_SOLVERS_KLOOPINSIDE to
165      ALLOW_AUTODIFF. This flag now needs to be set outside of      ALLOW_AUTODIFF. This flag now needs to be set outside of
166      solve_tri/pentadiagonal.F      solve_tri/pentadiagonal.F
Legend:
Removed from v.1.1517  
changed lines
  Added in v.1.1540
  ViewVC Help
Powered by ViewVC 1.1.22