This beta notes are divided into 5 part:
(A) Changes made in param.nml and other major inputs (as these are most affected when we hop versions)
(B) Changes in outputs
(C) Algorithmic changes and bug fixes
(D) Auto test history
(E) History for Pre-svn versions
================================================================================================
(A) Changes in param.[in,nml] and main parameter inputs for other modules:

Github versions:
a0960ed (25 Aug 2022): added 'iprecip_off_bnd' to optionally turn off precip near land bnd (to avoid some spurious plumes 
               near tropical islands etc);
ed26f60 (25 Aug 2022): removed 'impose_net_flux' and replaced it with IMPOSE_NET_FLUX (so this won't be actively supported);

Tag stofs3d-atl.v1.1.0 (identical to v5.10.0): for NOAA STOFS3D Atlantic operational forecast release
Tag v5.10.0: (13 July, 2022)

d03830e (16 May 2022): removed the 2 extra lines required in partition.prop (so it's identical to global_to_local.prop now);
a71ea69 (12 May 2022): removed 'ioffline_partition' and replaced it with CPP (NO_PARMETIS) to allow build without ParMETIS lib; renamed global_to_local.in as partition.prop;
da90a33 (2 May 2022): removed itr_met=1,2;
2165224 (29 April 2022): added optional offline partitioning 'ioffline_partition' to bypass ParMETIS; 
                         if on, needs global_to_local.in;
567b612 (28 April 2022): added new WWM outputs and re-ordered after iof_wwm(26); zCoordinate outputs made optional (for performance test);
6524e19 (27 April 2022): a few more WWM related parameters under VOR (turbinjds,alpha,wfwvor_streaming) 
                         and new WWM outputs, after merging with lrdev_updates;
d44dbdf (1 April, 2022): added an optional threshold inflation ratio for mass conversation for ICM: 'rinflation_icm'
6f367d0 (30 Mar 2022): added optional vertical levels to inject source tracer conventration: 'lev_tr_source'; 
7cc7f27 (30 Mar 2022): added an optional threshold depth for mass conversation for ICM: 'h_massconsv';
1788867 (16 Mar 2022): added optional parameters for S exchange (split from heat exchange): i_hmin_salt_ex,hmin_salt_ex;
dcd5356 (31 Jan 2022): removed ramp flags except for nramp_elev (use dramp*<=0 to turn off ramp): 
      nramp, nrampbc, nrampwind, nrampwafo,nramp_ss; removed the optional input 'shapiro_min.gr3';
80f1caa (26 Jan 2022): added 4 parameters in ice.nml: albsn,albsnm, albi,albm
d792ed0 (18 Jan 2022): added 2 new ice parameters: 'ice_atmos_stress_form' and 'cdwin0';
b4d2214 (7 Jan 2022): added a new parameter 'nu_sum_mult' to allow sum/product of final relax constants for tracer nudging;
f9f0844 (3 Jan 2022): ICE module, added 2 new parameters: lead_closing, Saterm. Also changed ice-wind stress formula;
7455d5d (30 Dec 2021): added a new option nws=-1 (PaHM, Parametric Hurricane Model);
91c91a7 (17 Dec 2021): ICE module, added new parameters: ncyc_fct, depth_ice_fct. Also requires ice_fct.gr3 now;
0e2f90e (17 Dec 2021): added a new parameter in ice.nml 'mevp_coef' and 2 related ones: mevp_alpha[34]
2a7647d (24&29 Nov 2021): added a new parmater 'loadtide_coef' for iloadtide=2,3;
a7a0b2b (17 Oct 2021): restored iwindoff;
afc7b1a (11 Oct 2021): changed vgrid.in format for ivcor=1 to facilitate parallel read (use
                       change_vgrid.f90 to convert) 
e9aec89 (1 Oct 2021): added most module outputs; order of flags changed in: ICE, ANALYSIS, SED;
85b585a (1 Sept 2021): add zcor output flag as iof_hydro(25); rest of hydro output flags (iof_hydro) shifted by 1;
8efc374 (30 Aug 2021): first version of scribe dictated I/O (as an option);
1902c1e (16 July 2021): changed hotstart.nc: added nsteps_from_cold and cumsum_eta for DA;

Tag v5.9.0: 657157a (1 July, 2021)

e281d94 (25 June 2021): added a new option for SAL (Stepanov & Hughes 2004): iloadtide=3;
bfb4afc (18 June, 2021): added an optional input 'shapiro_min.gr3' to be used with ishapiro=2;
d07d75d (April 21, 2021): added T,S in required inputs for offline transport (to use hydro only results);
b66e554 (April 14, 2021): added ishapiro=2 - Smagorinsky like filter option. In this option, shapiro0 is the coefficient;
b43afea (April 2, 2021): changed x,y to double in nc outputs (for newer visIT);
9150d86 (Mar 31, 2021): added a new SAL option (iloadtide=2) using scaling;
084e149 (Feb 25, 2021): Removed parameters: ibtrack_openbnd, iwindoff, dzb_decay (with hardwire); also
                        removed option for negative roughness. Fixed race condition for marsh module.
3576c9c (Jan 24, 2021): added new option for source/sink input: if_source=-1 requires source.nc (which
                        includes elem list inside; allows different time steps and # of records for
                        volume/mass source/sinks. Also, now the source/sink values in .th must be single
                        precision (not double);
bcdfce6 (Jan 6, 2021): added main switch for nc output 'nc_out' (useful for other programs to control outputs);
28ff9d1 (Dec 16, 2020): added optional self-attraction loading tides; if iloadtide/=0, need
                        amp/phases in loadtide_[FREQ].gr3 (freq's shared with tidal potential);
f8ba470 (Dec 3, 2020): Added a new option (meth_sink) to treat net sink: if an elem is dry with a net sink, vsource is reset
                       to 0;
f9043e2 (Nov. 12, 2020): merged with LRU branch wwm_lr; removed 'sav_cd' (isav=1 now requires sav_cd.gr3).
      New parameters related to WWM: fwvor_advxy_stokes,fwvor_advz_stokes,
      fwvor_advz_stokes,fwvor_breaking,cur_wwm,wafo_obcramp;
      Changes in WWM: ALPBJ ->B_ALP, BRHD->BRCR.
8083cfa (Nov. 10, 2020): added option for transport solver only ('itransport_only');
7e9ff0a (Aug 20, 2020): added two options (selected by 'i_hmin_airsea_ex') for locally turning off heat/salt exchange.
      i_hmin_airsea_ex=1: exchange turned off if local grid depth<hmin_airsea_ex
      i_hmin_airsea_ex=2: exchange turned off if local water depth<hmin_airsea_ex
      This replaces the change made by adde1aa (July 8, 2020): added a new parameter 'hmin_airsea_ex' (min total water depth for heat/salt exchange);
2f7bab1 (Jun 28,2020): reorganized ICM process (rates/fluxes) and added output flag in param.nml.

Tag v5.8.0: afd171c (June 14, 2020)

1477780 (May 8, 2020): changed sediment.in to sediment.nml (and made parameters lower case: BEDLOAD_COEFF, 
      NEWLAYER_THICK,IMETH_BED_EVOL, SAND_SD50=>Sd50, SAND_ERATE=>Erate, SED_TYPE=>iSedtype,
      SAND_SRHO=>Srho, SAND_WSED=>Wsed, SAND_TAU_CE=>tau_ce, sed_morph_fac=>morph_fac,
      poro_cst=>porosity);
aaf1306 (April 4, 2020): added hybrid ELM transport parameters: ielm_transport, max_subcyc;
eb00a00 (Mar 31, 2020): add 12th tracer model (USE_DVD of Klingbeil); if invoked, make sure: flag_ic(12)=0, and set b.c. flag and nudging option to 0;
cd9664a (Mar 16, 2020): [1st PR from Baptiste Mengual] added iof_sed(12:15) for new SED code. In sediment.in: 
                        new options for various parameters and a few new parameters;
                        sed_morph=1 requires imorphogrid.gr3 (a scale applied to depth change at each node);
                        also morp factors are replaced by a single 'sed_morph_fac' fro all classes;
73f7451 (Jan 14, 2020): changed ifltype=3 to be consistent with '5' (i.e., now the amp/phases should be variable along open bnd);
59741d4 (Jan 9, 2020): added another pre-proc flag 'ipre2' for more diagnostic outputs (can be used with >1 CPU; stops
                       after outputs);
4e2eb86 (Dec 11, 2019): added optional flags 'iadjust_mass_consv0' to enforce mass conservation for each of (11) tracer models; 
2648090 (Dec 1, 2019): added 'vclose_surf_frac' for correcting vertical vel/flux;
5ff2198 (Nov 17, 2019): added 'level_age(:)' if USE_AGE (AGE module revamped);
05b0099 (Sep 11, 2019): pre-pended in/out dirs, and all outputs are now in outputs/. Name changes: 
         fort.11-> fatal.error, flux.dat ->flux.out, total.dat->total.out, total_TR.dat->total_TR.out,
         fort.33->JCG.out, fort.17->subcycling.out;
f69ca44 (Sep 9, 2019): updated nc libs in GOTM3 make;
---------------------------------------------------------------
SVN versions:
R5217: replaced bottom friction coefficient output with bottom stress vector output;
R5204: added an option 'ibtrack_openbnd' - ibtrack_openbnd=0 recovers old results, but try '1';
R5191: fixed a bug in iwind_form;
R5183: Allowed junk values (-9999) in .nc under inu_[MOD]=2 - code will not nudge tracers 
       to those values. So 1 way to avoid nudging for a tracer is to set its nudged values to -9999;
R5180: add an option flag 'iunder_deep' and restored 'hw_depth' and 'hw_ratio' under iunder_deep/=0;
R5168: changed to param.nml (3 sections: CORE, OPT, SCHOUT). Renamed the following:
   shapiro ->shapiro0; ibcc ->ibc; itransport ->ibtp; cpp_lon ->slam0; cpp_lat ->sfea0; 
   bfric ->nchi; latitude ->rlatitude; coriolis ->coricoef; 
   turb_met ->mid (must use ''); turb_stab ->stab (must use ''); slvr_output_spool ->moitn0;
   mxitn ->mxitn0; tolerance ->rtol0; hotout ->nhot; hotout_write ->nhot_write;
   consv_check ->iflux; depth_zsigma ->h_bcc1; 
   ic_[MOD] -> flag_ic(:); inu_[MOD] ->inu_tr(:)
   All output flags (iof_ and iout_sta, nspool_sta,nhot,nhot_write) changed and moved 
   to SCHOUT section;
R5165: removed 'iPh' in icm.in (and replaced it with ICM_PH);
R5131: added 'impose_net_flux' (if nws=2 and impose_net_flux/=0, read in net surface heat flux as 'dlwrf' 
       (Downward Long Wave) in sflux_rad (solar radiation is still used separately), and if 
        PREC_EVAP is on, also read in net P-E as 'prate' (Surface Precipitation Rate) in sflux_prc);
R5125: added 'vnh[1,2] and vnf[1,2]' (vertical nudging); final relax is dt times the sum of horizontal and
       vertical relax. Use this e.g. to restore near surface;
R5123: added 'niter_shap' (# of iterations with Shapiro filter);
R5114: added 'i_prtnftl_weno' (option for writing nonfatal errors on invalid temp. or salinity for density);
R5049: added dbte and other outputs in ANALYSIS module;
R5032: added a new output (if ICM): ICM_DIN and ICM_PON
R5028: renamed sflux*.nc with an extra '0' to allow up to 9999 stacks (use sflux_rename.pl to rename)
R5017: added a new output (if ICM): ICM_PrmPrdt (primary production).
R5001: added a new output (if USE_SED): 'SED_TSC' (total suspended concentration).
       The corresponding global array is total_sus_conc(nvrt,npa) and can be used if USE_SED;
R4998: added iwind_form=1 (Hwang stress formulation);
R4970: WWM now requires hgrid_WWM.gr3 (same as hgrid.gr3 if pure triangular; 
       use split_quads_wwm.f90 to convert for mixed grids; no B.C. part needed). 
       Update 'FILEGRID' in .nml (GRID section);
R4947: WWM added 'MEVEG, LSOUBOUND, LACCEL'
R4904: changed nudging input format (inu_*=2) to reduce size (add_map_nudge.m to convert old);
R4892: added 'h[1,2]_bcc';
R4811: wwminput.nml (removed ISOURCE, LEXPORT_*, PARAMWRITE in &BOUC and a few others; renamed MDC, MSC);
R4772: added iwbl=2 (wave boundary layer formulation by Soulsby, 1997) (USE_WWM), c/o Kevin Martins;
R4697: added 'nrampwafo, drampwafo, turbinj' (matters only USE_WWM);
R4672: ics=2 can now work with ncor/=1;
R4644: nws=3 is now the option for coupling to atmos models;
R4636: added 'h_ml0','salt_ice','salt_water' in ice.nml;
R4634: added outputs 'ICE_net_heat_flux', 'ICE_fresh_water', 'ICE_top_T' (if USE_ICE);
R4582: added 'shorewafo' (only matters if USE_WWM; =0: original algorithm);
R4405: removed parameters: ihydlg, hw_depth, hw_ratio;
R4403: removed all '.6?' in output name list and appended '_side' and '_elem' for native outputs;
R4309: added 'gen_wsett' if USE_GEN for settling vel.
R4207: moved starting time info from sflux_inputs.txt and bctides.in to param.in. 
       sflux_inputs.txt now must NOT have this info (i.e. usually empty with only section headers);
       The 1st line of bctides.in is now for comments only.
       Changes in param.in: sim_[year,month,day]->start_[year,month,day] (integers); 
       removed sim_[minute,second]; 'start_hour' as double; added 'utc_start' (double);
R4116: changed all binary inputs to netcdf: *_[23]D.th =>*_[23]D.th.nc; [MOD]_nu.in =>[MOD]_nu.nc,
       hotstart.in => hotstart.nc. New version now needs netcdf4.4* and higher to compile 
       (also you only need -lnetcdff on some systems, not -lnetcdf).
R4025: added ICE module nc outputs (ICE_*) if USE_ICE
R4009: reinstated 'nramp_elev' (for ihot/=0)
R3976: added 'nstep_ice' (coupling with ice) if USE_ICE
R3699: removed 'depo_scale' and 'relath' in sediment.in;
R3636: new SAV code. Inputs for isav=1: sav_[h,D,N].gr3;
R3630; removed 'idrag';
R3619: SAV model added with 2 new parameters: isav and sav_cd;
R3563: removed 'inflow_mth' and output flags for Tsinghua group with USE_SED;
R3554: moved to newer svn repo versin 1.9.3  (columbia);
R3540: added PetSc as an option; no changes in param.in yet;
R3526: moved 2-phase mixture parameters to sediment.in; only 'inflow_mth' is retaind;
R3457: optional parameters from Tsinghua group: Two_phase_mix and output flags etc. Needed if USE_SED is on; 
R3399: added 'inflow_mth' (0: default uniform vel as flow b.c.);
R3362: moved 'rho0' and 'shw' to param.in;
R3286: removed 'nramp_elev' (not needed);
R3227: a new parameter 'slr_rate' (SLR rate) if USE_MARSH;
R3209: added a marsh migration model (inputs: marsh_init.prop, marsh_barrier.prop). New parameters:
       'mrsh.66';
R3098: removed 'nonhydro', 'ihydro_region';
R3087: added type '5' b.c. for elev. and vel (combination of tides and *[23]D.th). For
       vel, also '-5' is similar to '-4'; 
R3084: added two new parameters for under resolution in a b-clinic model:
       'hw_depth' and 'hw_ratio' (Hannah-Wright-like ratio & depth to detect under resolution);
R3080: added 'ieos_pres' (/=0: add hydrostatic pressure effects in EOS);
R3073: added a new flow b.c. (ifltype=-2, iettype=0) for discharge-stage relation. If invoked, need
       to input 4 coefficients of the polynomial function in bctides.in;
R3062: sediment.in: added erosional formulation options in SED3D (ierosion=0: Ariathurai & Arulanandan);
R3050: added a new paramter 'ihydlg' for hydrology;
R3017: added 2 options in SED2D (multi-class only): 'ISED_DUMP' (dumping/dredging), and 'MORPH_FAC' (morph acceleration);
R2956: hvis_coef0 now used for ihorcon=1,2 (no hvis_coef.gr3); 'shapiro' only needed if ishapiro=1;
       added ishapiro=-1 option (with shapiro.gr3) for transitioning between eddying and non-eddying regimes;
R2953: sediment.in: added dumping/dredging option in SED3D (ised_dump);
R2947: quads added in SED3D;
R2936: restored Shapiro filter as an option: 'ishapiro' and 'shapiro'. This option is normally used
       with indvel=ihorcon=0;
R2935: new parameters, 'rearth_pole', 'rearth_eq', 'iupwind_mom', 'hvis_coef0'; removed 'cdh'. In addition, 
       ihorcon=2 for bi-harmonic viscosity; hvis coefficients are specified in hvis_coef.gr3
       if ihorcon=1, and hvis_coef0 if ihorcon=2;
R2875: sediment.in, added a new parameter 'relath';
R2863: add nws=5,6 options (inputs: UVP.nc or UVP_direct.nc);
R2830: reinstate 'xlsc0' (mixing length scale for b.c.) as a parameter in param.in;
R2820: sediment.in, a parameter 'depo_scale' (scaling for depositional flux);
R2800: sediment.in, replaced 'bc_for_weno' by 'ised_bc_bot' (1: Warner's bottom b.c.; 2: Tsinghua Univ). 
       Hard-coded 3 parameters: g,vonKar,rhom;
R2796: exposed tolerances for implicit TVD in param.in: 'eps1_tvd_imp' and 'eps2_tvd_imp';
R2792: added 'ieos_type' for type of Eq. of State used (0: original UNESCO; 1: linear function of T ONLY). If
       ieos_type=1, needs two new parameters: eos_[a,b] for linear function;
R2789: added 'sed_morph_time' in sediment.in;
R2768: restored Manning's friction option (bfric=-1) and related 'hmin_man'. Note that if you use this
       option, Manning's formula will be applied to ALL prisms including 3D ones;
R2687: (1) 'ntracers' replaced by that for each tracer model: ntracer_gen, ntracer_age, sed_class, eco_class 
           (for ICM it's hardwired to 23);
       (2) icst repaced by 'ic_TEM' and 'ic_SAL' (and they must be same); 
           i.c. flags for other tracer models are ic_[MOD]. The inputs for T,S are not changed (ts.ic etc),
           but for other tracers, it's [MOD]_hvar_[ID].ic or [MOD]_vvar_[ID].ic, where [ID] are local tracer #;
       (3) the transport method for all tracers is given by 'itr_met' (which replaces iupwind_t);
       (4) nudging option flags are now: inu_[MOD] (including T,S). The relax. inputs are: [MOD]_nudge.gr3.
           Type '2' inputs are [MOD]_nu.in, and step is 'step_nu_tr';
       (5) Output name changes for other tracers: [MOD]_[ID].6?. In general, model name is added to the old names
           (see param.in for other names);
       (5) removed: flag_model,flag_ic, iupwind_t,inu_st, inu_tr, step_nu, ntracers
       (6) changes in names of .th: [MOD]_[ID].th (for all tracers and type 1 b.c.; temp.th->TEM_1.th, 
           salt.th->SAL_1.th); [MOD]_3D.th (includes all local tracers in it for type 4 b.c., but all of these
           must use same time step!), and so
           temp3D.th->TEM_3D.th, salt3D.th->SAL_3D.th. msource.th should include _all_ tracers if enabled
           (e.g. GEN,  AGE);
       (7) In addition formats are changed for bctides.in and hotstart, if tracer modules are enabled (use combine_hotstart6.f90);
R2650: removed 'flimiter' and 'flimiter2' - TVD options need to be set in Makefile.local now;
R2648: removed 'tvd_mid*' (option for TVD formulations);
R2581: indvel=0 now works with quad/tri. The only remaining non-working option is inunfl=1;
R2559: removed 'inter_st' (ELM transport);
R2552: added ramp-up flags for source/sinks (nramp_ss and dramp_ss) if if_source=1;
R2549: removed ELM transport option (iupwind_t=0);
R2266: new format from quads2/. Obsolete parameters are listed near the end. Also quads does not work
       with indvel<=0. vgrid.in is
       now required even for 2D model (1 layer). Have not worked on any of other modules yet;
       Changed parameters: (1) bfric=-1 is removed (for 2D sides, log formula is used which assumes the total depth as the BBL);

R2237: depth.61 output redefined as the _change_ in depth from initial condition (i.e. hgrid.gr3);
R2208: add a new output file 'bdrc.61';
R2040: format change for msource.th and wind.th (nws=1,4);
R18xx: added 2 new outputs from SED3D: bthk.66 (total bed thickness in meters) and
       bage.66 (total bed ages in sec);
R1842: add GOTM4.0 in make. More tests are needed.
R1801: added a new parameter for tracer nudging: step_nu_tr (for inu_tr=2, which requires an input tr_nu.in); also ICM sed. outputs: bdoc.66,bnh4.66,bno3.66,bpo4.66,bcod.66,sbdo.66,sbsa.66;
R1790: added a new option for momentum bottom b.c. ibottom_bc=1 recovers old results; =2: solve
       mom. eq. from surface to true bottom.
R1781: added age calculation. New outputs are age_[1..N].63. Method: set ntracers=2*N, where
       N is # of age tracers (the tracer array is [c_1..c_N a_1..a_N], and age_i=a_i/c_i). 
       All i.c. =0; c_i=1, c_j=0 (j/=i), a_[1..N]=0 at relevant bnd(s), and itrtype=0 at ocean bnd;
R1699: .th now can have its own time step, but MUST start at t=0! Previously only wind.th can have its own dt;
R1587: added a new parameter: ishapiro_violation. If =0, no Shapiro filter violation check will be done.
R1577: (not param.in per se) added a few new parameters in vgrid.in for ivcor (types of z-coord. systems; 2 for SZ)
R1534: (merged with branch/selfe_opt1) tvd_mid, tvd_mid2, flimiter, flimiter2 should only be changed if CHOOSE_TVD is turned on;
R1495: added a new parameter (dzb_decay<=0) for friction option bfric=1.
       Under this option, Cd is calculated from log law when dzb>=dzb_min, and 
       Cdmax=Cd(dzb=dzb_min). When dzb<dzb_min, Cd=Cdmax*exp[dzb_decay*(1-dzb/dzb_min)].
       e.g, if dzb_decay=-3.4539, Cd(dzb=dzb_min/3)=0.1*Cdmax. Note that the old
       bfric=1 option can be recovered by setting dzb_decay=0;
R1460: corrected mis-spelt 'latitude';
R1405: 1 new parameter (hmin_radstress) used only for rad. stress calculation;
R1382: relaxed check for rmaxvel;
R1359: removed 'ishapiro'; it's set internally instead;
R1354: 1 new parameter (cdh) needed when ihorcon/=0 - land friction coeeficient.
R1350: 1 new parameter (ishapiro) to do pre-proc for Shapiro filter.
R1343: 2 new outputs for SED2D: cflsed.61 and qav.62
R1341: 2 new parameters: inv_atm_bnd, prmsl_ref (inverse barometric effect option for elev. b.c.)
R1302: Moved tracer parameters from bctides.in to param.in: itr_met,tvd_mid2,flimiter2,inu_tr
R1282: output bfmt rename as z0st; 
R1272: streamlined non-standard output section. All of these (bfmt,z0eq,z0cr,z0sw,z0wr,bpgr,wafo) 
       must be present in param.in;
R1089: removed mass_source
R1038: added new parameter cdsed.61 for SED2D
R849: remove obsolete parameter iwrite;
R841: added a new parameter 'mass_source' to bypass mass source in transport;
R812: ics=2 and inter_mom>0 works now;
R747: replacd Cdmax with dzb_min for bfric=1 (roughness). With dzb_min=0.5m, the max Cd's are: 0.0022 (z0=0.1mm), 
      0.0034 (z0=0.5mm), 0.0041 (z0=1mm), 0.0061 (z0=3mm), 0.0075 (z0=5mm), 0.0105 (z0=1cm);
R744: added SED2D output parameters;
R513: add 2D Manning formulation inside code; bfric=-1 is the new option (needs hmin_man and manning.gr3). 2D model
      must use this option.
R473: new parameter: if_source (for pt source/sink);
R348: new parameter: irouse_test. Also mdc2,msc2,iwbl, iharind, flag_ic, flag_model and sim_* now must be present (even when not used);
R300: 'hvel.65'-> 'hvel.67', 'vert.66'->'vert.69', 'temp.67'->'temp.70','salt.67'->'salt.70';
      added non-standard output option for internal arrays for generic use;
R292: added nramp_elev;
R250: removed 'testout'; added 'hvel.65', 'vert.66','temp.67', 'salt.67'
R192: msc, mdc changed to msc2, mdc2 
R180: added msc, mdc if WWM is used (same as MSC,MDC in .nml);
R158: new parameter: ihydraulics

Changes in wwminput.nml (WWM):
R4861: MSC, MDC, MESIN, LLIMT, LIMFAK restored
R3797: MSC, MDC changed to NUMSIG, NUMDIR; MESIN changed to ISOURCE (on/off); MELIM default is 1 (on). 
       removed: LLIMT, LIMFAK. Add 2 new sections: SIN4, SDS4;
R326: (1) &OUTP renamed &STATION; added &HISTORY
      (2) LMONO_IN, LMONO_OUT, LNAUTOUT moved to &PROC
      (3) LWXFN moved to &HISTORY
R192: (1) DMIN now linked to h0 in param.in;
      (2) &BOUC: GRIDTYPE changed to IGRIDTYPE (3: SELFE); BOUNDFORMAT chanegd to IBOUNDFORMAT;
      (3) &BOUC: added LPARMDIR (1D spectra only)
      (4) &WIND: LWINDFROMWWM may be used to read wind into WWM directly;
      (5) &ENGS: FRICC must >0
R??: (1) FRHIG --> FRHIGH
     (2) added LVECTOR, IVECTOR in NUM section (Sept. 11, 2011);

---------------------------------------------------------------
Changes in other inputs:
(20) R2684: bctides: combined T,S and tracers b.c. flags into 1 line;
(19) R2570: removed xlsc.gr3 as input (itur=3), and set it at 0.1;
(18) R2266: vgrid.in is now required for 2D model;
(17) R2040: make all .th format consistent
            msource.th: all tracers are lumped into 1 line [ath3(1:nsources,1:2+ntracers,...)]; 
            wind.th (nws=1,4): add time stamp;
(16) R1850: SED3D input changes: (a) removed bedthick_overall from sediment.in and 
                                 added bedthick.ic (.gr3 format); (b)in addition, also need bed_frac_*.ic.
(15) R1813: renamed vvd.dat as vvd.in;
(14) R1699: .th now can have its own time step, but MUST start at t=0! Previously only wind.th can have its own dt;
(13) R1302: Moved tracer parameters from bctides.in to param.in: itr_met,tvd_mid2,flimiter2,inu_tr
(12) R1264:  SED2D part added to hotstart. Use combine_hotstart5.f90 for combining.
             hotstart.in is backward compatible for pure Hydro (i.e. no ICM or SED2D);
(11) R1225: changed tvd.gr3 and fluxflag.gr3 to .prop format;
(10) R1222: appled the change in R1219 to elev. and vel. nudging as well;
(9) R1219: changed relax. const. definition (suggested by Ben and Brett). The values in 
           [s,t,tracer]_nudg.gr3 will be multiplied by dt to get the final relax. const.
           One approach to calculate the values in these .gr3 files is: -log(0.1)/T90/86400, where
           T90 is the 90% decay time in days;
(8) R1019: fluxflag.gr3 now takes an arbitrary # of regions from -1,0...M. The code
           (with consv_check=1) will calculate the flux between 2 region numbers with difference=1, excluding region -1;
(7) R934: If running on Stampede, need an input 'die.stam' (temp. fix for cluster hang problem);
(6) R835: fluxflag.gr3 can now take an arbitrary # of regions from 0...M. The code
          (with consv_check=1) will calculate the lower triangle of the fluxes matrix. 
(5) R513: if bfric=-1 (2D model), needs manning.gr3 (Manning's n);
(4) R473: if if_source=1, need: source_sink.in, vsource.th, vsink.th, 
          and msource.th (volume sources and sinks are combined in the
          code, which is then ramped up with a rate 2x slower than main ramp if nramp=1; 
          no ramp for mass source);
(3) R3??: elev3D.th renamed as elev2D.th;
(2) R86: tvd.gr3 is needed if iupwind_t=2;
(1) R??: if ihydraulics/=0, need hydraulics.in 

================================================================================================
(B) Changes in outputs:
(1) R??: 3D profiles for station outputs;
(2) R147: commented out (1);
(3) R304: added non-standard outputs (*.67 etc) that users can also use.
          For side- or center-based outputs, combine_output6 needs sidecenters.gr3
          or centers.gr3;
(4) R881: add timer_comp() for generic use in compution-only portion of a routine for load balance analysis etc.
(5) R964: hotstart outputs changed to include additional arrays for ICM; hotstart.in is 
          backward compatible if USE_ICM is not on. Use combine_hotstart4.f90 for combining.
(6) R1469; consolidated many output files to reduce I/O footprint;
(7) R1776: hotstart outputs changed to include SED[23]D arrays; use combine_hotstart5.f90 for combining.

GIT versions:
(1) e227f8579 (Dec 2020): added wet/dry flags @nodes/sides in nc outputs;

================================================================================================
(C) Algorithmic changes & bug fixes:
(1) R97: found a bug in evaluate_cubic_spline
(2) R212: changes in btrack routines to enhance robustness
(3) R395: broke main routine into _init, _step, and _finalize
(4) R752: fixed a bug in iwind_form
(5) R812: added and verified kriging ELM in lat/lon (ics=2)
(6) R840: fixed a bug in hotstart for station output due to change in output format;
(7) R854: added a new flag (imet_dry) in hgrad_nodes, to deal with wet/dry on a wet internal side. 
          Right now imet_dry=2 is only used for radiation stress, but may consider expand its use to others
          (b-clinic, non-hydro pressure).
(8) R883: ICM added to svn.
(9) R1032: qel(np) to match solver dimension.
(10) R1060: fixed a bug related to radiation stress in lon/lat (wrong frame).
(11) R1071: change in itur=3: solve closure eq. from nvrt to kbp+1 (following Tuomas K.'s suggestion).
(12) R1076: fixed a bug in b.c. for GOTM closure; seem to make small differences in general.
(13) R1096: fixed a bug in tvd.gr3 (should be node based);
(14) R1296: Kijin fixed a bug in hydraulic struc;
(15) R1322: deallocate all temp. arrays in routine to avoid memory leak.
(16) R1338: put non-standard output routine to a module; this is important on some systems like IBM. 
(17) R1354: restored land friction option (that works in conjunction with horizontal viscosity).
(18) R1534 (branch/selfe_opt1): swapped indices of the following arrays (some also changed name):
     ic3
     ssign
     js -> elside
     nm -> elnode
     is() -> isdel
     isidenode
     jsgb
     inp -> indnd
     ine -> indel
     iself
     iselfgb
     nmgb
     ic3gb
     cspline_ypp_nd
     cspline_ypp_sd
     dl -> dldxy
     isidenei2
     sparsem

     Also some arrays in main code only: sne,itier_nd,akrmat_nd, df[v,h,q1,q2],xl,q2,[st]nd_nu[1,2,-],isidenei2
(19) R1577: added ivcor=1 option.
(20) R1603: first release for new ivcor=1 option (localized sigma and degenerated prisms); not done yet.
(21) R1641: ivcor=1 done, including some post-proc FORTRAN scripts. ACE: slab at a z does not work; rest OK; m-elio: will not work
for ivcor=1. 
(22) R1672: changed the flag in the call to area_coord() in quicksearch() to enhance robustness;
            also updated v3.2.1.
(23) R1677: Fixed bugs related to swapping of array mcoefd() indices in SED[23]D; also updated v3.2.1. 
(24) R1776: Fixed bugs related to swapping of array in SED3D (avalanching part);
(25) R1798: ICM; removed the changed in w-vel by YC. Tested by ZG on MCB and Upper Bay;
(26) R2039: removed the 3D profile part of station outputs;
(27) R2557: fixed a bug in water type with mixed tri/quads.
(27) R2266: from branches/quads2/, with new mixed tri/quads, and 2/3D prisms.
(28) R2294: gone thru modules (except WWM) and fixed TVD with quads (2-tier ghosts).
(29) R2406: removed OHSU stuff eventually;
(30) R2549: removed ELM transport (iupwind_t=0) and associated arrays;
(31) R2583: fixed a bug in 2-tier ghost table;
(32) R2696: in the i.c part, removed 'Impose no slip b.c. to be consistent with ELM transport';
(33) R2715: fixed a bug in implicit transport related to the divisor for other terms like 
            diffusion etc;
(34) R2726: fixed a few bugs related to tracer 3D.th and nudging inputs;
(35) R2731: changed implicit transport to 3-step splitting to take care of b.c. from vertical diffusion part;
(36) R2940: removed Shapiro filter and changed horizontal viscosity to be like a filter. Added bi-harmonic viscosity.
            First version of upwind for mom. advection;
(37) R2983: fixed a bug in imp transport (spatial limiter part) c/o Fei.
(38) R2992: fixed a bug in inunfl=1 (incompatible assignment)
(39) R3022: fixed a bug for 2D model (bcc not init'ed)
(40) R3065: fixed a bug in combine_output8 (nc output);
(41) R3098: removed non-hydro option;
(42) R3101: streamlined ics=2; all vectors are in local lon/lat frame including s[uv]2 now;
(43) R3153: fixed a bug for ics=2 and quads;
(44) R3207: revert some parts of _step (from R3111);
(45) R3273: fixed a bug in ifltype=-2 b.c. (from v5.3);
(46) R3294: reverse a loop in bktrk_subs (NASA's suggestion);
(46b) R3299: experimented with 3rd order Adam Bathforth scheme for Coriolis: results are similar.
(47) R3317: a bug related to SED3D hotstart (wrong dim);
(49) R3349: array index bugs in COSINE fixed (introduced in R3342);
(50) R3386: merged dev branch of openMP back to trunk; not completely done yet;
(51) R3444: replace rho0 in buoyancy freq with prho();
(52) R3496: tweak I/O for Fujitsu
(53) R3540: added PetSc as an option
(54) R3580: fixed a bug related to ibdef that affects imm=1
(55) R3619: SAV model added
(56) R3630: fixed a bug with quads&imm/=0;
(57) R3636: new SAV code based on theory of Shimizu and Tsujimoto (1994);
(58) R3678: fixed a bug related to Winterwerp formulation (SED3D);
(59) R3699: reworked depositonal flux in SED3D (D=w_s*C; removed depo_scale); 
(60) R3703: fixed a bug related to morph acceleration: the bottom b.c. for transport should 
            not be scaled by MF;
(61) R3797: new WWM-III source code, LLIMT was removed and ISOURCE was added for simplification of the source terms switches
       WWM-III supports now WWIII, WAM and SWAN physics, where ST4 from WWIII is the default physics as used at NCEP/NOAA and ECMWF in the future. 
       The whole code was rewritten in terms efficiency, like chache locality + arrays allignement has been improved and so the scalibility of the code
       ASPAR_LOCAL from 1-10 was introduced to have influence of memory locality, e.g. for ASPAR_LOCAL_LEVEL = 1 as much as possible is precomputed 
       and using ASPAR_LOCAL_LEVEL = 10 u have basically everything computed on the fly. What the best is for the certain system and problem can be tested 
       by try and error for small computer systems and smaller problems ASPAR_LOCAL_LEVEL = 1-4 and for large scale problems 5-10. 
(62) R3932: Hydro/lap.F90, line 1663 has Mbound issue. Aron fixed it by declaring D[XY](N) in DCOPY and DDOT.
(63) R3945: switch to netcdf outputs; saved last binary output version as branches/v5.4
(64) R3976: added a new option for settling vel. (Richard Hofmeister).
(65) R4009: fixed a bug with ihot=2 introduced between R3285 and R3286 due to the removal of nramp_elev.
     Without it the results won't match. Reinstated nramp_elev and applied same fix to v5.4.
(66) R4116: (from Hai) changed the deposition mass calculation wrt age in SED3D
(67) R4118: finished netcdf related changes to I/O and scripts
(68) R4212: reset tempmin to -5C for ice module.
(69) R4273: added evap&precip into volume source.
(70) R4309: implicit treatment for vertical migration in itr_met>2
(71) R4407: bug fixes related to vortex force (wet/dry; index)
(72) R4500: revamped sediment flux module inside ICM
(73) R4549: Changed ptrack3.f90 behavior when a particle hits horizontal bnd or wet/dry bnd - 
            it'll now be reflected back as in LTRAN. This avoids particles being stuck
(74) R4644: nws=3 is now option for coupling to atmos models
(75) R4699: Carten Lemmon changed ParMetis-3.1-Sep2010/METISLib/stdheaders.h and ParMetis-3.1-Sep2010/ParMETISLib/stdheaders.h
        (added __clang__)
(76) R4729: Aron Roland changed ParMetis-3.1-Sep2010/Makefile.in and added '-D...':
     OPTFLAGS = -O2 -D_Float128=__float128
(77) R4812: fixed a bug in wwm_bdcons.F90 (array VAL out of bound)
(78) R4815: fixd a bug in SED3D output (bed_fracn)
(79) R4825: fixd a bug in compute_average3.f90 (Nicole Cai)
(80) R4831: removed w_s* in btrack in vortex force
(81) R4861: WWM reverted to R3012; trunk version moved to branch;
(82) R4947: analysis module added
(83) R4968: quads added for WWM
(84) R4984: branches/coriolis3 experimented using node vel to compute Coriolis to reduce
            inertial instability. Didn't seem to make much difference.
(85) R4994: allows volume sources/sinks to be specified more than once at same element 
            (code will accumulate). The corresponding mass source concentration 
            will take one of the specified values (no summation).
(86) R5005: Aron fixed an issue with shallow-water sources in WWM
(87) R5028: allow up to 9999 stacks for sflux*.nc
(88) R5044: fixed a bug in ptrack3 to deal with junk values like below bottom
(89) R5068: fixed a bug in init of rough_p for USE_SED (dry spots)
(90) R5079: added option to work with uncombined nc outputs in some post-proc FORTRAN & mlab scripts
(91) R5178: fix bug of saturate DO in ICM
(92) R5187: incorporated dry bnd treatment from bndry (c/o Jens Wyrwa). However, to get 
            good volume conservation it's best to keep all open flow bnd's wet all the time as before;
(93) R5191: fixed a bug in iwind_form (accidentally re-init'ed after reading);
(94) R5202: fixed a bug on negative ICM tracer concentration - check negative values before sending 
            back to hydro.
(95) R5204: a new option to control behavior of btrack when trajectory hots an open bnd 
            (set vel=0 and exit). Goal is to eventually use this as default;

git versions:
(96) 64b7181: fixed an efficiency issue and added nc checks in ptrack3 (c/o Marcel Rieker)
(97) 2138e77: Fixed a major bug introduced during reshuffling of _init for PDAF: need to call nodavel before sflux, as (uu2,vv20 are needed there.
(98) 5ff2198: revamped AGE module. B.C. now all should use 0. Only 0 and 1 should be used in AGE_hvar_[1:ntr/2].ic, 
              where ntr is @ of age tracers, and the code will 'hold' concentration at 1 at those prisms 
              (level specified in param.nml) that have 1 as i.c. I.C. in AGE_hvar_[ntr/2+1:ntr].ic should =0.
              Injecting '1' at dry elem's is allowed but results may be harder to interpret.
(99) 6cec698: bug fix for spherical coord in WWM (around dateline);
(100) c46c7bd (Feb 14, 2020): Fei tweaked WENO solver (wrt upwind);
(101) d3adb2c (Feb 14, 2020): changed to bilinear interp for HYCOM hot and nudging scripts (to speed up);
(102) 7506147 (Mar 1, 2020): add limiter for settling vel to avoid char line out of bnd;
(103) 5665418 (Mar 18, 2020): PR from Baptiste mengual for SED3D (sed_frition.F90 merged after the rest).
(104) eb00a00 (Mar 31, 2020): added 12th tracer model (USE_DVD of Klingbeil)
(105) aaf1306 (April 4, 2020): added hybrid ELM transport for efficiency (via branch hybrid_ELM);
(106) 40955ab (April 16, 2020): replaced the fatal error in nadv=2 (aptivity trap) with exit
(107) e703189 (May 6, 2020): reverted to old simple approach for deposional mass (c/o Baptiste Mengual);
(108) 14e201f (May 7, 2020): switched to ParMETIS v4.0.3;
(109) 33fb37f (Jun 30, 2020): fixed a bug in ielm_transport (Fei Y.).
(110) 55910d1 (July 31, 2020; via branch mem_leak_nc4): reverted the change in schism_io (close/open 
               nc output after/b4 each write, to accommodate PDAF's flexible mode), as this has caused a 
               lot of mem leaking (discovered by Nicole C. when testing ICM).
(111) eaf9093 (Aug 2020): Baptiste M. modified the computation of the bottom shear stress 
              under combined waves and currents in SED3D
(112) 4ac56df (Aug 2020): Paul Ryan and Claire Trenham (CSIRO) fixed some init array issues in hydro and WWM.
(113) 7e9ff0a (Aug 20, 2020): Fei Y fixed a msource bug (init of msource at 0 at elem's not in source_sink.in
                              led to ice rain;
(114) f311026 (Aug. 27, 2020): reverted init of msource to 0 for tracers other than T,S, 
      c/o Nicole Cai. Using ambient values for other tracers can lead to artificial and additional accumulation of nutrients e.g.

(115) 00b6f16 (Nov. 9, 2020): PR #21 from Kijin merged (SED): changed method to compute near bottom vel for LSC2.
(116) 8083cfa (Nov. 10, 2020): added option for transport solver only (itransport_only)
(117) 6dc44d3 (Dec 14, 2020): Fixed a bug, thanks to Kevin Kartin, on vortex formulism of wave force (missing a 
                              factor of area() in I_4);
(118) 28ff9d1 (Dec 16, 2020): added optional self-attraction loading tides. The option shares some constants
                              with tidal potential: freq names and cut-off depth;
(119) b1bcaa0 (April 20, 2021): removed most of 'goto'. Remaining ones: harm.F90, lap.F90, WWM
(120) 0cec024 (April 21, 2021): bug fix in misc_subs c/o of Fei (inunfl=1: nodeA in final extrap stage may be interface node)
(121) fb30239 (April 22, 2021): bug fix for ellipsoidal earth (tensor frames).
(122) fb79cdc (May 26, 2021): in interpolate_depth_structured2*, added an option to shift 1/2 cell for ll corner
                              (c/o Charles Seaton);
(123) 5e87c24 (June 9, 2021): more changes in interpolate_depth_structured2* to extrap also into right/upper sides.
(124) 843c40f (19 June, 2021): fixed a bug in ptrack3 (pt_in_poly3; c/o Jilian Xiong) that affects quads;
(124) b2cf92b (29 June, 2021): fixed a bug in station outputs in basin scale cases ics=2 (local proj is not accurate
                               if the station is far away from the local frame);
(125) a9f8f6c (8 July, 2021): fixed a bug in iloadtide=1 (index error). Also revamped reading of input files
                              for large core counts.
(126) ab46def (20 July, 2021): merged multi ice model (CICE) from Qian Wang.
(127) 70db927 (11 Aug 2021): bug fixes from Qian Wang on ice matrix (ghost nodes)
(128) 8efc374 (30 Aug 2021): first version of scribe dictated I/O (as an option)
(129) 53689f4 (12 Sept 2021): fixed a bug in weno transport introduced in e794d8c (23 Aug 2021).
(130) c7dd516 (27 Sept 2021): AGE module tweaked (clamped @ i.c. at injection points for both concentrations).
(131) (19 Oct 2021): filtered isolated wet nodes during outputs (should not change other results);
(132) 7455d5d (30 Dec 2021): added a new option nws=-1 (PaHM, Parametric Hurricane Model);
(133) 7455d5d (30 Dec 2021): added Fariall air-sea exchange (USE_BULK_FAIRALL) c/o Jerome Lefevre;
(134) e288f9a (31 Jan 2022): removed the restriction of counter-clockwise orientation in sflux grid;
(135) dde345f (7 April 2022): fixed a bug in sflux, invoked only if nws=2, ihconsv=1, and iwind_form=0: 
                              uninit'ed stress values.
(136) 6524e19 (27 April 2022): merged with La Rochelle group on WWM  (VOR);
(137) cd0b2a3 (14 Aug 2022): [via Qian_pole_treatment] changed nodalvel() under ics=2; commented out a fram change
                             for better vel near poles.
================================================================================================
(D) Auto-test history:
R100 (Nov 2011); R168 (minus WWM); R224 (basic); R240 (full); R306 (basic); R370 (basic); R408 (basic); R430 (basic); R601 (basic); R730 (basic); R747 (all hydro); R1120 (hydro+ SF BayDelta); R1305: (all hydro; ICM); R1532 (branch/selfe_opt1): all; R1641: hydro; R2018: hydro; R2232 (all);
(SCHISM) R2521: all modules tested;
R2737: all modules except SED3D tested;
R4899: SCHISM+WWM tested

ICM notes:
(1) Point source input (ps.in) right now has a hard-wired dt of 1 day (starting from 0).
(2) R1147: a) nps (# of point sources) is now read in from wqparam.in (17th line with 16th being comment);
           b) renamed pselfe_sed_test_inp as icm_sed.in; removed COLDI and HOTI in 
              this file, and use ihot instead;

---------------------------------------------------------------
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Tag (tag) releases:
v5.8.0: afd171c (June 14, 2020)
v5.7.0 (R5132)
v5.6.1 (R4542)
v5.5.0 (R4191)
v5.4.1 (R4748)
v5.3.1 (R3228?)
v3.2.0 (R242): 
(1) Hydro and WWM fully tested; ready for operational use;
(2) hydraulic option not ready;

v3.4.0 (R2264, Nov. 2014): all modules tested.

Release branch:
v5.7 (R5146)
v4.1 (R2403): eventually removed OHSU stuff;
v4.1 (with changes from OHSU):
v3.2.1 (R1700)
v3.3.0 (from R1873/trunk)
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
gnu make system:

To use Makefile, three files are needed:
mk/Make.defs.local, mk/Make.defs.config, and src/Makefile
The file mk/Make.defs.local should be written depending on different environment. 
Write your own env into mk/Make.defs.<your env name> (see mk/Make.defs.* from others as a reference),
 and copy or symlink it to mk/Make.defs.local. svn ci the first file (not the 2nd file). You can then toggle on/off
 modules using mk/Make.defs.local.

In addition, python is also needed.

To add a new CPP or module, need to update all 3 files (in the order shown above).
================================================================================================
(E) Pre-svn versions:
v3.0b:

v3.0c:
       (5) param.in:
           a) now include all parameters in tracer_param.in if ntracers/=0;
           b) if nonhydro=1, need ihydro_region to indicate if a hydrostatic region needs to
               be specified in hydro_region.gr3 (depth=1 in hydrostatic region);
           c) added station output option (iout_sta; if /=0 need nspool_sta and an input 'station.in');
           d) if USE_HA is turned on in Makefile, 'iharind' is needed to do harmonic analysis;
           e) new option imm=2;
           f) if NAPZD is used, 2 extra output flags: Bbdf.63 and totN.63;

v3.0d:
       (3) param.in:
           a) if WWM is used, add a coupling flag (icou_elfe_wwm=0: decoupled so 2 models
              will run independently); also an output flag 'Hsig.61';

v3.1a:
       (5) param.in:
          a) im2d added (1: 2D model);
          b) no long reset zcor.63 output flag;
          c) some parameters no longer matter for 2D model (see sample);
          d) 'Hsig.61' output added if USE_WWM;
          e) ics=2: sphericall coordinate option;

v3.1b:
       (2) param.in:
          a) add 'theta2' - implicitness factor for Coriolis (2D only);
          b) add 'btrack_noise': a scale to avoid underflow in btrack (for init. nudging);
          c) WWM: output 'WavD.61' added;

v3.1c:

v3.1d: 
       (3) bctides.in:
           a) nudging factor needed for all b.c. for T,S, and tracers (except '0'); 
              type -1 & -4 removed (use type 3 or 4);
           b) tracer transport: b.c. -1 replaced by 3 (nudging to i.c.);
       (9) param.in:
           a) nstep_wwm; new icou_elfe_wwm=2;

v3.1dc: v3.1db with changes in v3.1kj in btrack.
        (1) param.in:
           a) dtb_max[1,2] replaced by dtb_max, dtb_min (max/min dt for all schemes);
           b) btrack_noise removed (hardwired in the code).

v3.1g: 
       (4) param.in:
           a) a new parameter "izonal5" to turn on Williamson test #5 (zonal flow over 
              an isolated mount)

v3.1h: 
       (2) hotstart for station outputs should work now;
       (3) bctides.in: tracer b.c. type 1 (input htr_?.th where "?" is tracer #) 
           and 4 (input tr3D.th) added; all types (except for "0") need nudging factor;
       (6) param.in:
           a) a new parameter iwind_form (=-1 with nws>=2 uses old Pond formulation as
              nws=1; =0: use the value calculated from the heat exchange routine if nws>=2);
           b) new WWM output vars.

V3.1i: 
       (2) param.in:
           a) a new parameter h_tvd if iupwind_t=2 (upwind is used if h<h_tvd);
           b) btrack_noise replaced by btrack_nudge (default=1.e-2);


V3.1j: 
       (2) param.in:
          a) ibtrack_test added (=1: rotating Gausshill test with stratified T,S);

V3.1k: 
       (2) param.in:
          a) 2 new flags (inu_elev and inu_uv); if they=1, need inputs 
             elev_nudge.gr3 or uv_nudge.gr3;

V3.1ke: 
        (3) param.in:
           a) vnf1,vnf2=1 to keep horizontal relax;

V3.1kf: 
        (1) interpol.gr3 removed; interpolation in SZ region changed;

V3.1kg: V3.1kf with 
        (2) param.in:
           a) new parameter iwbl for WBL (modified Grant-Madsen formulation);

V3.1kj:
        (3) param.in:
           a) dtb_max[1,2] replaced by dtb_max, dtb_min (max/min dt for all
schemes);
           b) nws=4 option: ascii wind.th (uwind,vwind, pres. at all nodes at
each time);
           c) re-organized WWM outputs;