--- MITgcm/doc/tag-index	2015/06/09 16:31:45	1.2157
+++ MITgcm/doc/tag-index	2015/10/22 15:51:59	1.2191
@@ -1,9 +1,186 @@
-$Header: /home/ubuntu/mnt/e9_copy/MITgcm/doc/tag-index,v 1.2157 2015/06/09 16:31:45 rpa Exp $
+$Header: /home/ubuntu/mnt/e9_copy/MITgcm/doc/tag-index,v 1.2191 2015/10/22 15:51:59 mlosch Exp $
 $Name:  $
 
     Notes on tags used in MITgcmUV
     ==============================
 
+o pkg/seaice/seaice_evp.F: fix store directives and key computations, since
+  there is no adjoint test for EVP, we don't know what it does to the adjoint
+  solution, but at least there are no more recomputation warnings left
+o pkg/seaice/seaice_evp.F: slight modification of averaging so that EVP stands
+  a chance of giving the same results as a fully converged VP (JFNK) solver
+o tools/adjoint_options/adjoint_diva:
+  - fix bug (typo) in Tang-Linear FLAGS setting for MPI + DIVA
+o pkg/streamice:
+  - spotted bug in streamice_driving_stress using dyG where should have been dxG
+o pkg/exf:
+  - remove STOP statement that prevented to use wind or (wind-)stress
+    interpolation for constant-in-time forcing fields (wind/stress period = 0).
+o pkg/seaice:
+  - replace SEAICE_EVP_USE_ADAPTATION by run time parameters, requires
+    5 additionals 2D fields and 6 new store directives
+  - change logic in seaice_readparms.F: EVP code can now be turned on in
+    various ways, setting SEAICE_deltaTev is no longer required,
+    alternatively set SEAICE_evpAlpha/Beta, SEAICEuseEVPstar, SEAICEuseEVPrev,
+    or SEAICEaEVPcoeff
+o pkg/shelfice/shelfice_thermodynamics.F
+  - redefining uLoc/vLoc to u/v, instead of 2u/v
+o pkg/seaice/seaice_evp.F: introduce "adaptive EVP" by Madlen Kimmritz
+  - this is an intermediate state of a truely converging EVP that has
+    the potential of outperforming implicit VP methods, I do not recommend
+    using it yet.
+  - for now, "adaptive EVP" is turned on by setting SEAICE_EVP_USE_ADAPTATION
+    and all parameters are hardwired
+  - for convergence it still requires very smooth regularisations, that means,
+    if SEAICE_DELTA_SMOOTHREG is defined, deltaCreg=deltaC+deltaMin, which is
+    different from the VP case with deltaCreg=sqrt(deltaC**2+deltaMin**2); also
+    specific averaging appears to be required, still subject to tuning
+  - still needs cleaning and I need to turn parameter into runtime parameters
+  - add code to compute and print residuals of iteration, compile with
+    defining ALLOW_SEAICE_EVP_RESIDUAL to enable it
+o pkg/seaice/seaice_evp.F: fix a bug (factor of four in front of e12Csq)
+  that entered with r1.54 on Sep04, 2015, -> update results
+
+checkpoint65o (2015/09/14)
+o pkg/mom_vecinv & mom_common:
+  - add deep-atmosphere geometry factor in mom_calc_visc.F + in routines
+    called from there.
+  - add anelastic and deep-atmosphere geometry factor in pkg/mom_vecinv ; this
+    allows to use Vector-Invariant form in deep atmos and anelastic formulation
+o pkg/seaice: introduce fast ice parameterisation following Itkin et al. (2015)
+  and Koenig-Beatty+Holland (2012)
+  - two new parameters SEAICE_tensilFac and SEAICE_tensilDepth
+  - global 2D field tensileStrength, computed in seaice_calc_ice_strength
+  - requires an extra input argument for S/R seaice_calc_viscosities
+  - tensileStrength's depth dependence is different from Itkin et al. (2015)
+    (to be changed and played with, for now exponential)
+o pkg/seaice: introduce new parameter SEAICEscaleSurfStress (default = .FALSE.)
+  - if TRUE scale ice-ocean and ice-atmosphere stress acting on sea ice
+    by AREA according to Connelly et al. (2004), JPO.
+  - For EVP and the LSR solver, the implicit part of the drag term is
+    the only non-zero term in the denominator when there is no ice (which
+    prevented running the model with zero ice-ocean drag). If
+    SEAICEscaleSurfStress = .TRUE., this results in a division by zero
+    (or zero main diagonals BU/BV) which need to be caught. A practical
+    consequence is that for open water, the momentum equation reduce
+    to trivially 0 = 0 + 0 (for EVP). For LSR, BU/BV are reset to 1, if
+    they happen to be zero, often leading to a non-zero solution over
+    open water.
+o pkg/seaice: revise EVP code (seaice_evp.F) in preparation for a more
+    efficient EVP method.
+  - introduce area-weighted averages for e12 (analogously to
+    seaice_calc_viscosities.F); this totally changes lab_sea.hb87 (2 digits
+    remain, but also changing the order of terms in the previous average
+    changed the results as much) -> update experiment
+  - add SEAICE_DELTA_SMOOTHREG code
+  - disentangle computation of zetaC/deltaC and zetaZ/deltaZ, also for
+    TEM-code
+  - adjust loop bounds so that only required fields are computed
+  - add new averaging code for zetaZ/deltaZ (again similar to
+    seaice_calc_viscosities.F), inactive by default for now (will change
+    results again), but will replace old code eventually
+o eesupp/src:
+  - add an other alternative for multi-proc GLOBAL_SUM (in global_sum_tile.F)
+    to always cumulate tile local-sum in the same order
+    (option #define GLOBAL_SUM_ORDER_TILES in CPP_EEOPTIONS.h).
+    Still uses MPI_Allreduce (like standard code) but on array of tiles
+    and should be faster than GLOBAL_SUM_SEND_RECV code.
+  - switch almost all verification experiments that use GLOBAL_SUM_SEND_RECV
+    (defined in CPP_EEOPTIONS.h_mpi) to GLOBAL_SUM_ORDER_TILES instead.
+  - add CPP_EEOPTIONS.h_mpi with GLOBAL_SUM_ORDER_TILES defined in 14 more exp.:
+     aim.5l_cs, cheapAML_box, deep_anelastic, dome, fizhi-cs-32x32x40,
+     fizhi-cs-aqualev20, fizhi-gridalt-hs, global_ocean_ebm, matrix_example,
+     natl_box, solid-body.cs-32x32x1, tutorial_advection_in_gyre,
+     tutorial_baroclinic_gyre and tutorial_held_suarez_cs.
+
+checkpoint65n (2015/07/29)
+o pkg/thsice:
+  - add a warning when used with pkg/aim but without updating SST
+    (stepFwd_oceMxL=F & useCoupler=F & tauRelax_MxL<>-1 ).
+o pkg/offline:
+  - change where Offline-Fields are loaded, from load_fields_driver.F to
+    directly within FORWARD_STEP, just after updating iter number and time;
+  - move setting of GM and KPP diffusivity out of OFFLINE_FIELDS_LOAD into new
+    S/R OFFLINE_GET_DIFFUS that is called towards the end of DO_OCEANIC_PHYS;
+   These changes allow:
+    a) to recover exactly on-line results using deltaToffline=deltaTClock,
+    tested (using offlineIter0= nIter0 -1 and offlineTimeOffset= -deltaT/2)
+    both with Lin & Non-Lin Free-Surf, both with staggerTimeStep & synchronous
+    time-step, both with GM & KPP files and with recomputing these fields;
+    b) no needs to switch on staggerTimeStep anymore;
+    c) simplify rescaling of offline horiz. velocity with Non-Lin Free-Surf.
+  - update output of experiment tutorial_dic_adjoffline (only 3 digits left).
+o pkg/offline:
+  - change (rarely used) offset-iteration parameter to an offset time (in s):
+    can easily recover previous offset-iter effect by setting offset-time
+    (offlineTimeOffset) to offset-iter x deltaToffline;
+  - do not turn-off Temp & Salinity surface relaxation when using KPP (used
+    in KPP as part of surface buoyancy forcing) in offline run.
+  - fix ptracers KPP non-local term in the case where KPP is re-computed.
+o OpenAD
+  - Adding preliminary ggl90 test to OpenAD verification.
+    Config. with ggl90 currently diverges after 64 time steps
+    so will need more work to become more stable
+  - Adding preliminary kpp test to OpenAD verification.
+    required small tweakes in kpp_routines.F
+  - Merge and update from Krishna Narayanan's contrib area:
+    * Add one variable for preliminary OpenAD DIVA
+    * Add replacement time-stepping for OpenAD DIVA
+    * genmake2 flag -diva (but only for OpenAD; preliminary)
+    * required modifs for OAD_support
+    successful preliminary test for tutorial_global_oce_biogeo/
+o pkg/seaice:
+  - initialise deltaC in seaice_lsr, cosmetic changes in seaice_lsr
+  - add cpp-flag SEAICE_DELTA_SMOOTHREG for option of regularising
+    delta with a smooth function in s/r seaice_calc_viscosity
+    (no effect on EVP)
+  - rename local suffix "sqr" into "sq" for more consistent variable names
+o pkg/seaice:
+    add new parameter SEAICE_deltaMin that is used ***only***
+    for regularizing Delta (and nothing else, like the parameter
+    which is also used for all sorts of things SEAICE_EPS).
+    Defaults to SEAICE_EPS for backward compatibility only.
+o pkg/seaice: replace erroneously storing e12Csqr on the wrong tape with
+    the wrong key by inialising it before use -> fixes a recomputation
+    problem but does not fix the gradient
+o pkg/layers:
+  - move resetting of flux arrays to outside the iLa loop
+    (this caused layers_diapycnal to not work with more than one layer coord.)
+  - fix bug in calculation of layers_recip_delta in layers_init_fixed.F
+o pkg/layers:
+  - remove from common block 3D layers fields (used to diagnose transport
+    in tracer coords) and declare them as local arrays in layers_calc.F
+o pkg/seaice:
+  - Rename SEAICE_VECTORIZE_LSR_ZEBRA to SEAICE_LSR_ZEBRA and define it
+    in lab_sea forward experiment (changes results -> updated)
+  - update adjoint experiment lab_sea after cleaning up seaice_lsr_tridiagu/v
+o pkg/seaice: modify seaice_lsr.F in order to improve the gradient
+   computations; for SEAICE_VECTORIZE_LSR
+  - move the loops over which the tridiagonal solvers (seaice_lsr_tridiagu/v)
+    sweep to the outside,
+  - remove store directives and add "CADJ loop sequential" directives
+    in analogy to model/src/solve_tridiagonal.F
+  - replace many "#ifdef SEAICE_VECTORIZE_LSR_ZEBRA" by variable loop
+    boundaries to yield more readable code. This has also the charming side
+    effect that your can use SEAICE_VECTORIZE_LSR_ZEBRA without
+    SEAICE_VECTORIZE_LSR (i.e. adjoint without recomputations in
+    seaice_lsr_tridiagu/v now requires either of these flags, vectorization
+    still requires SEAICE_VECTORIZE_LSR with SEAICE_VECTORIZE_LSR_ZEBRA as
+    an additional option)
+   The tridiagonal solvers are now completely analoguous to solve_tridiagonal.F
+   but the adjoint gradients (which are affected by this change) still explode.
+o coupler (atm & ocn _compon_interf):
+  - avoid using Exch2 IO mapping if not suitable.
+
+checkpoint65m (2015/06/15)
+o pkg/layers:
+  - clean-up old parameters setting (from Sep 2012):
+    + remove old parameters setting from common block (declared locally);
+    + stop if any old parameter is mixed with new parameters;
+    + check for valid or consistent parameter setting (e.g., layers_name)
+  - check that layers_bounds for Pot.Density are in the right range;
+    this is supposed to catch un-updated setting after recently shifting
+    potential density field "prho" by -1000 (now in "sigma" scale).
 o pkg/layers:
   - Added missing call to layers_fill for keeping track of tottend in
     pkg/diagnostics/diagnostics_fill_state.F
@@ -21,8 +198,8 @@
     thermodynamics.F, impldiff.F) and gad to save the necessary fields
     for layers_thermodynamics calculations
 o pkg/ptracers:
-  - add capability of stepping a passive tracer forward 
-    without advection by setting PTRACERS_advScheme to 0.
+  - add capability of stepping a passive tracer forward without advection
+    by setting PTRACERS_advScheme to 0.
 o pkg/diagnostics:
   - add few (threads) _BARRIER around any update of "diag_pkgStatus".
 o pkg/layers: