First commit

README.NMM

@@ -0,0 +1,292 @@
+
+WRF-NMM Model Version 3.2 (March 31, 2010)
+
+----------------------------
+WRF-NMM PUBLIC DOMAIN NOTICE
+----------------------------
+
+WRF-NMM was developed at National Centers for
+Environmental Prediction (NCEP), which is part of
+NOAA's National Weather Service.  As a government
+entity, NCEP makes no proprietary claims, either
+statutory or otherwise, to this version and release of
+WRF-NMM and consider WRF-NMM to be in the public
+domain for use by any person or entity for any purpose
+without any fee or charge. NCEP requests that any WRF
+user include this notice on any partial or full copies
+of WRF-NMM. WRF-NMM is provided on an "AS IS" basis
+and any warranties, either express or implied,
+including but not limited to implied warranties of
+non-infringement, originality, merchantability and
+fitness for a particular purpose, are disclaimed. In
+no event shall NOAA, NWS or NCEP be liable for any
+damages, whatsoever, whether direct, indirect,
+consequential or special, that arise out of or in
+connection with the access, use or performance of
+WRF-NMM, including infringement actions.
+
+================================================
+
+V3 Release Notes:
+-----------------
+
+This is the main directory for the WRF Version 3 source code release.
+
+- For directions on compiling WRF for NMM, see below or the 
+  WRF-NMM Users' Web page (http://www.dtcenter.org/wrf-nmm/users/)
+- Read the README.namelist file in the run/ directory (or on 
+  the WRF-NMM Users' page), and make changes carefully.
+
+For questions, send mail to [email protected]
+
+Release Notes:
+-------------------
+
+Version 3.2 is released on March 31, 2010.
+
+- For more information on WRF V3.2 release, visit WRF-NMM Users home page
+  http://www.dtcenter.org/wrf-nmm/users/, and read the online User's Guide.
+- WRF V3 executable will work with V3.1 wrfinput/wrfbdy.  As 
+  always, rerunning the new programs is recommended.
+
+The Online User's Guide has also been updated.
+================================================
+
+The ./compile script at the top level allows for easy selection of 
+NMM and ARW cores of WRF at compile time. 
+
+   - Specify your WRF-NMM option by setting the appropriate environment variable:
+
+         setenv WRF_NMM_CORE 1
+           setenv WRF_NMM_NEST 1 (if nesting capability is desired)
+         setenv HWRF 1 (if HWRF coupling/physics are desired)
+
+   - The Registry files for NMM and ARW are not integrated
+     yet. There are separate versions:
+
+         Registry/Registry.NMM         <-- for NMM
+	   Registry/Registry.NMM_NEST    <-- for NMM with nesting
+         Registry/Registry.EM          <-- for ARW (formerly known as Eulerian Mass)
+
+
+How to configure, compile and run?
+----------------------------------
+
+- In WRFV3 directory, type:
+
+   configure
+
+  this will create a configure.wrf file that has appropriate compile 
+  options for the supported computers. Edit your configure.wrf file as needed.
+
+  Note: WRF requires netCDF library. If your netCDF library is installed in
+        some odd directory, set environment variable NETCDF before you type
+        'configure'. For example: 
+
+        setenv NETCDF /usr/local/lib32/r4i4
+
+- Type: 
+        compile nmm_real
+
+- If sucessful, this command will create nmm_real.exe and wrf.exe
+  in directory main/, and the appropriate executables will be linked into
+  the run directories under test/nmm_real, or run/.
+
+- cd to the appropriate test or run directory to run "nmm_real.exe" and "wrf.exe".
+
+- Place files from WPS (met_nmm.*, geo_nmm_nest*)
+  in the appropriate directory, type
+
+  real_nmm.exe
+
+  to produce wrfbdy_d01 and wrfinput_d01. Then type
+
+  wrf.exe
+
+  to run.
+
+- If you use mpich, type
+
+  mpirun -np number-of-processors wrf.exe
+
+=============================================================================
+
+What is in WRF-NMM V3.2?
+
+* Dynamics:
+
+  - The WRF-NMM model is a fully compressible, non-hydrostatic model with a
+    hydrostatic option.
+
+  - Supports One-way and two-way static and moving nests.
+
+  - The terrain following hybrid pressure sigma vertical coordinate is used.
+
+  - The grid staggering is the Arakawa E-grid.
+
+  - The same time step is used for all terms.
+
+  - Time stepping:
+     - Horizontally propagating fast-waves: Forward-backward scheme
+     - Veryically propagating sound waves: Implicit scheme
+
+  - Advection (time):
+     T,U,V:
+      - Horizontal: The Adams-Bashforth scheme
+      - Vertical:   The Crank-Nicholson scheme
+     TKE, water species: Forward, flux-corrected (called every two timesteps)/Eulerian, Adams-Bashforth
+     and Crank-Nicholson with monotonization.
+
+  - Advection (space):
+     T,U,V:
+      - Horizontal: Energy and enstrophy conserving,
+        quadratic conservative,second order
+
+      - Vertical: Quadratic conservative,second order, implicit
+
+      - Tracers (water species and TKE): upstream, positive definite, conservative antifiltering
+        gradient restoration, optional, see next bullet.
+
+      - Tracers (water species, TKE, and test tracer rrw): Eulerian with monotonization, coupled with
+        continuity equation, conservative, positive definite, monotone, optional.  To turn on/off, set
+        the logical switch "euler" in solve_nmm.F to .true./.false.  The monotonization parameter
+        steep in subroutine mono should be in the range 0.96-1.0.  For most natural tracers steep=1.
+        should be adequate.  Smaller values of steep are recommended for idealizaed tests with very
+        steep gradients.  This option is available only with Ferrier microphysics.
+
+  - Horizontal diffusion: Forward, second order "Smagorinsky-type"
+
+  - Vertical Diffusion:
+     See "Free atmosphere turbulence above surface layer" section
+     in "Physics" section given in below.
+
+  - Added a new highly-conservative passive advection scheme to v3.2
+
+Added Operational Hurricane WRF (HWRF) components to v3.2. These enhancements include:
+     - Vortex following moving nest for NMM
+     - Ocean coupling (with POM)
+     - Changes in diffusion coefficients
+     - Modifications/additions to physics schemes (tuned for the tropics)
+         - Updated existing SAS cumulus scheme
+         - Updated existing GFS boundary layer scheme
+         - Added new HWRF microphysics scheme         - Added new HWRF radiation scheme
+     Please see the WRF for Hurricanes webpage for more details:
+         http://www.dtcenter.org/HurrWRF/users
+
+
+* Physics:
+
+  - Explicit Microphysics: WRF Single Moment 5 and 6 class /
+    Ferrier (Used operationally at NCEP.) / Thompson [a new version in 3.1] 
+    / HWRF microphysics: (Used operationally at NCEP for HWRF)
+
+  - Cumulus parameterization: Kain-Fritsch with shallow convection /
+    Betts-Miller-Janjic (Used operationally at NCEP.)/ Grell-Devenyi ensemble
+    / Simplified Arakawa-Schubert (Used operationally at NCEP for HWRF)
+
+  - Free atmosphere turbulence above surface layer: Mellor-Yamada-Janjic (Used operationally at NCEP.)
+
+  - Planetary boundary layer: YSU /  Mellor-Yamada-Janjic (Used operationally at NCEP.) 
+    / NCEP Global Forecast System scheme (Used operationally at NCEP for HWRF)
+    / GFS / Quasi-Normal Scale Elimination
+
+  - Surface layer: Similarity theory scheme with viscous sublayers
+    over both solid surfaces and water points (Janjic - Used operatinally at NCEP).
+    / GFS / YSU / Quasi-Normal Scale Elimination / GFDL surface layer (Used operationally at NCEP for HWRF)
+
+  - Soil model: Noah land-surface model (4-level - Used operationally at NCEP) /
+    RUC LSM (6-level) / GFDL slab model (Used operationally at NCEP for HWRF)
+
+  - Radiation:
+    - Longwave radiation: GFDL Scheme  (Fels-Schwarzkopf) (Used 
+      operationally at NCEP.) / Modified GFDL scheme (Used operationally 
+      at NCEP for HWRF) / RRTM
+    - Shortwave radiation: GFDL-scheme (Lacis-Hansen) (Used operationally 
+      at NCEP.) / Modified GFDL shortwave (Used operationally at NCEP 
+      for HWRF)/ Dudhia
+
+  - Gravity wave drag with mountain wave blocking (Alpert; Kim and Arakawa)
+
+  - Sea Surface temperature updates during long simulations
+
+* WRF Software:
+
+  - Hierarchical software architecture that insulates scientific code
+    (Model Layer) from computer architecture (Driver Layer)
+  -  Multi-level parallelism supporting distributed-memory (MPI)
+  -  Active data registry: defines and manages model state fields, I/O,
+    nesting, configuration, and numerous other aspects of WRF through a single file,
+    called the Registry
+  - Two-way nesting:
+      Easy to extend: forcing and feedback of new fields specified by
+        editing a single table in the Registry
+      Efficient: 5-8% overhead on 64 processes of IBM
+  - Enhanced I/O options:
+      NetCDF and Parallel HDF5 formats
+      Nine auxiliary input and history output streams separately controllable through the
+       namelist
+      Output file names and time-stamps specifiable through namelist
+  -  Efficient execution on a range of computing platforms:
+      IBM SP systems, (e.g. NCAR "bluevista","blueice","bluefire" Power5-based system)
+      IBM Blue Gene
+      SGI Origin and Altix
+      Linux/Intel
+         IA64 MPP (HP Superdome, SGI Altix, NCSA Teragrid systems)
+         IA64 SMP
+         x86_64 (e.g. TACC's "Ranger", NOAA/GSD "wJet" )
+         PGI, Intel, Pathscale, gfortran, g95 compilers supported
+      Sun Solaris (single threaded and SMP)
+      Cray X1, X1e (vector), XT3/4 (Opteron)
+      Mac Intel/ppc, PGI/ifort/g95
+      NEC SX/8
+      HP-UX
+      Fujitsu VPP 5000
+  - RSL_LITE: communication layer, scalable to very large domains, supports nesting.
+  - I/O: NetCDF, parallel NetCDF (Argonne), HDF5, GRIB, raw binary, Quilting (asynchronous I/O)
+, MCEL (coupling)
+  - ESMF Time Management, including exact arithmetic for fractional
+    time steps (no drift).
+  - ESMF integration - WRF can be run as an ESMF component.
+  -  Improved documentation, both on-line (web based browsing tools) and in-line
+
+    (Model Layer) from computer architecture (Driver Layer)
+  - Multi-level parallelism supporting shared-memory (OpenMP), distributed-memory (MPI), 
+    and hybrid share/distributed modes of execution
+  - Serial compilation can be used for single-domain runs but not for runs with
+    nesting at this time.
+  - Active data registry: defines and manages model state fields, I/O,
+    configuration, and numerous other aspects of WRF through a single file, 
+    called the Registry
+  - Enhanced I/O options:
+      NetCDF and Parallel HDF5 formats
+      Five auxiliary history output streams separately controllable through the namelist
+      Output file names and time-stamps specifiable through namelist
+
+  - Testing: Various regression tests are performed on HP/Compaq systems at
+    NCAR/MMM whenever a change is introduced into WRF cores. 
+
+  - Efficient execution on a range of computing platforms:
+      IBM SP systems, (e.g. NCAR "bluevista","blueice" and NCEP's "blue", Power4-based system)
+      HP/Compaq Alpha/OSF workstation, SMP, and MPP systems (e.g. Pittsburgh 
+         Supercomputing Center TCS)
+      SGI Origin and Altix
+      Linux/Intel
+         IA64 MPP (HP Superdome, SGI Altix, NCSA Teragrid systems)
+         IA64 SMP
+         Pentium 3/4 SMP and SMP clusters (NOAA/FSL iJet system)
+      PGI and Intel compilers supported
+      Alpha Linux (NOAA/FSL Jet system)
+      Sun Solaris (single threaded and SMP)
+      Cray X1
+      HP-UX
+      Other ports under development:
+         NEC SX/6
+         Fujitsu VPP 5000
+  - RSL_LITE: communication layer, scalable to very
+    large domains 
+  - ESMF Time Management, including exact arithmetic for fractional
+    time steps (no drift); model start, stop, run length and I/O frequencies are
+    now specified as times and time intervals 
+  - Improved documentation, both on-line (web based browsing tools) and in-line
+
+--------------------------------------------------------------------------

README.NSSL_AIA

@@ -0,0 +1,60 @@
+This is a brief document to the changes associated with 
+adding L Wicker's adaptive implicit advection.  Soon
+to be on the NY Times best selling list..
+
+Code changes are confined to:
+
+./Registry/Registry.EM_COMMON
+
+where I added two variables to the "dynamics" namelist
+
+rconfig   real    w_crit_cfl              namelist,dynamics 1             2.0     irh    "w_crit_cfl"            ""critical W-CFL where w-damping is applied""
+rconfig   integer zadvect_implicit        namelist,dynamics 1             0       irh    "zadvect_implicit"      ""turns on adaptive implicit advection in vertical""
+
+The first variable is used to specify the W-CFL value to relax to when w-damping is turned on.  I used this for RK5 implementations - default value is same as compiled value (2.0)
+The second variable is where the magic is happening - setting zadvect_implicit (default is off) equal to "1", turns on the adaptive implicit advection.
+
+./dyn_em
+
+I made substantial changes to 4 modules:
+
+a)   Module_em.F
+     Initial changes to module_em.F merged with WRF_AIA code module_em.F
+
+b)   Solve_em.F
+     Added changes from AIA code to solve_em.F  RK4 and RK5 integrations schemes are now implemented.
+
+c)   Module_advect_em.F
+     Added semi-lagrangian changes to the scalar_pd and wenopd routines.  These will be useful at any time, since the vertical courant number can be larger than 1 and SL upstream is PD.
+
+d)   Module_big_step_utilities_em.F 
+     added ww split code for ex/im advection, and updated w_damp routine to be more flexible with using w_crit_cfl from namelist
+
+The only place I messed with HRRR-specific codes is where I had added (caused I needed the height information) the "phi" and "phib" arrays already in some calling interfaces,
+and the HRRR code had already added those arrays.  The HRRR code put the extra arrays further down in the argument list, and I had them all with the other 3D arrays.  I think
+I did that in 2-3 places, but in all cases, the arrays were passed, simply in a different location in the list.  So you might run into that issue with any merger with other code.
+
+Added subroutines into the modules are:
+
+ww_split [Module_big_step_utilities_em.F]
+advect_u/v/w/phi/s_implicit, TRIDIAG, TRIDIAG2D [module_advect_em.F]
+
+TRIDIAG is not used - if you can be clever about speeding up TRIDIA2D, that will gain you time, as it is called every large time step for all advection variables and each column.
+The implicit advecton is called only on the last RK step.
+
+Note about implicit_advection routines:
+  I have formulated the implicit solution in terms of increments to the tendency arrays - so that roundoff should not be a problem.  Also, to make sure, I used double precision
+  variables locally in the routines and the TRIDIAG2 solution.  I have tried to run both ways, and clearly, SP words will be faster, although I am thinking that a lot of the
+  temporary arrays, which are 2D in (i,k), may be in cache even using DP words.
+
+Note about the increased use of "rk_order" in the code:
+  In older versions of WRF, you would run across a lot of statements, for the WENO and PD calls which are "rk_step < 3".  In order to implement the RK4, RK5, and AIA seamlessly,
+  I have made sure that "< 3" is now "< rk_order", which is extracted at the top of subroutines where it is used.  So in the namelist, rk_ord can be 2, 3, 4, 5
+
+Note about extra printing:
+  I have some extra diagnostic printing that is outputted ONLY ONCE at the beginning of the run.  If you need to turn all off, look for parameter statements at the top of
+  solve_em, ww_split and w_damp for the print_flag and set to false.
+
+
+Lou Wicker, Dec 13th, 2018
+...with modifications by J. Kenyon (8 Feb 2019) for compatibility with the hybrid vertical coordinate