# halo HALO_NMM_K dyn_nmm8:q2;24:t,u,v,q,w,zThe keyword is halo. The communication is named in the column, sothat it can be referenced in the source code. The entry in the column iscase sensitive (the convention is to start the name with HALO_NMM). The selecteddynamical core is defined in the column. There is no ambiguity, as everycommunication in each Registry file will have the exact same column option.The last set of information is the . The portion in front of the “:” is thestencil size, and the comma-separated list afterwards defines the variables that arecommunicated with that stencil size. Different stencil sizes are available, and are “;”separated in the same column. The stencil sizes 8, 24, 48 all refer to asquare with an odd number of grid cells on a side, with the center grid cell removed (8 =3x3-1, 24 = 5x5-1, 48 = 7x7-1). The special small stencil 4 is just a simple north, south,east, west communication pattern.The convention in the <strong>WRF</strong> model is to provide a communication immediately after avariable has been updated. The communications are restricted to the mediation layer (anintermediate layer of the software that is placed between the framework level and themodel level. The model level is where developers spend most of their time. Themajority of users will insert communications into the dyn_nmm/solve_m,m.Fsubroutine. The HALO_NMM_K communication defined in the Registry file, in theexample above, is activated by inserting a small section of code that includes anautomatically generated code segment into the solve routine, via standard cpp directives.#ifdef DM_PARALLEL# include "HALO_NMM_K.inc"#endifThe parallel communications are only required when the <strong>WRF</strong> code is built fordistributed-memory parallel processing, which accounts for the surrounding #ifdef.The period communications are required when periodic lateral boundary conditions areselected (ARW only). The Registry syntax is very similar for period and halocommunications, but the stencil size refers to how many grid cells to communicate, in adirection that is normal to the periodic boundary.# period PERIOD_EM_COUPLE_A dyn_em 2:mub,mu_1,mu_2The xpose (a data transpose) entry is used when decomposed data is to be re-decomposed(ARW only). This is required when doing FFTs in the x-direction for polar filtering, forexample. No stencil size is necessary.<strong>WRF</strong>-NMM V3: User’s Guide 6-11
# xpose XPOSE_POLAR_FILTER_T dyn_em t_2,t_xxx,dum_yyyIt is anticipated that many users will add to the the parallel communications portion of theRegistry file (halo and period. It is unlikely that users will add xpose fields.Registry PackageThe package option in the Registry file associates fields with particular physicspackages. Presently, it is mandatory that all 4-D arrays be assigned. Any 4-D array thatis not associated with the selected physics option at run-time is not allocated, used for IO,or communicated. All other 2-D and 3-D arrays are eligible for use with a packageassignment, but that is not required.The purpose of the package option is to allow users to reduce the memory used by themodel, since only “necessary” fields are processed. An example for a microphysicsscheme is given below.# package kesslerscheme mp_physics==1 - moist:qv,qc,qrThe entry keyword is package, and is associated with the single physics option listedunder . The package is referenced in the code in Fortran IF andCASE statements by the name given in the column, instead of the moreconfusing and typical IF ( mp_physics == 1 ) approach. The column muststart with a dash character and then a blank “- “ (for historical reasons of backwardcompatibility). The syntax of the column then is a 4-D array name,followed by a colon, and then a comma-separated list of the 3-D arrays constituting that4-D amalgamation. In the example above, the 4-D array is moist, and the selected 3-Darrays are qv, qc, and qr. If more than one 4-D array is required, a “;” separates thosesections from each other in the column.In addition to handling 4-D arrays and their underlying component 3-D arrays, thepackage entry is able to associate generic state variables, as shown in the examplefollowing. If the namelist variable use_wps_input is set to 1, then the variables u_gc andv_gc are available to be processed.# package realonly use_wps_input==1 - state:u_gc,v_gcI/O Applications Program Interface (I/O API)The software that implements <strong>WRF</strong> I/O, like the software that implements the model ingeneral, is organized hierarchically, as a “software stack”<strong>WRF</strong>-NMM V3: User’s Guide 6-12
- Page 1 and 2:
ForewordUser's Guide for the NMM Co
- Page 3 and 4:
• WPP Directory Structure 7-3•
- Page 5 and 6:
The WRF modeling system software is
- Page 7 and 8:
WRF-NMM FLOW CHARTTerrestrialDataMo
- Page 9 and 10:
Vendor Hardware OS CompilerCray X1
- Page 11 and 12:
If all of these executables are def
- Page 13 and 14:
Once the tar file is obtained, gunz
- Page 15 and 16:
HWRF is set, then (3) will be autom
- Page 17 and 18:
In addition to these three links, a
- Page 19 and 20:
• Real-data simulations• Non-hy
- Page 21 and 22:
k. Morrison double-moment scheme (1
- Page 23 and 24:
g. GFDL surface layer (88): (This s
- Page 25 and 26:
Other physics optionsa. gwd_opt: Gr
- Page 27 and 28:
to the convergence of meridians app
- Page 29 and 30:
Variable NamesValue(Example)Descrip
- Page 31 and 32:
Variable NamesValue(Example)Descrip
- Page 33 and 34:
Variable NamesValue(Example)Descrip
- Page 35 and 36:
Variable NamesValue(Example)Descrip
- Page 37 and 38:
Variable NamesValue(Example)Descrip
- Page 39 and 40:
mpirun.lsf wrf.exeand for interacti
- Page 41 and 42:
The boundary conditions for the nes
- Page 43 and 44:
Examples:1. One nest and one level
- Page 45 and 46:
ottom_top_stag = 28 ;soil_layers_st
- Page 47 and 48:
float HLENSW(Time, south_north, wes
- Page 49 and 50:
float HBOTS(Time, south_north, west
- Page 51 and 52: operational mesoscale Eta model. J.
- Page 53 and 54: Mlawer, E. J., S. J. Taubman, P. D.
- Page 55 and 56: NCEP WRF Postprocessor (WPP)WPP Int
- Page 57 and 58: ./configureYou will be given a list
- Page 59 and 60: equested output field. If the pre-r
- Page 61 and 62: RAINCV SNOW HBOTRAINNCVSNOWCNote: F
- Page 63 and 64: Running WPPFour scripts for running
- Page 65 and 66: i. As the grid id of a pre-defined
- Page 67 and 68: The GrADS package is available from
- Page 69 and 70: Height on pressure surface HEIGHT O
- Page 71 and 72: MELTPrecipitation type (4 types) -
- Page 73 and 74: Press at tropopause PRESS AT TROPOP
- Page 75 and 76: RIP4RIP IntroductionRIP (which stan
- Page 77 and 78: A successful compilation will resul
- Page 79 and 80: iinterp = 0v v v v vH V H V h h h h
- Page 81 and 82: espectively, of the centered domain
- Page 83 and 84: This can be either 'h' (for hours),
- Page 85: eal variable expect values that are
- Page 88 and 89: of all the requested trajectories a
- Page 90 and 91: Creating Vis5D Dataset with RIPVis5
- Page 92 and 93: User’s Guide for the NMM Core of
- Page 94 and 95: Configuration: The configure script
- Page 96 and 97: This program reads the contents of
- Page 98 and 99: Period - Describes communications f
- Page 100 and 101: fine grid), f (forcing, how the lat
- Page 104 and 105: (http://www.mmm.ucar.edu/wrf/WG2/Ti
- Page 106 and 107: User's Guide for the NMM Core of th
- Page 108 and 109: 2. Make sure the files listed below
- Page 110 and 111: User's Guide for the NMM Core of th
- Page 112 and 113: Program geogridThe purpose of geogr
- Page 114 and 115: 1. Define a model coarse domain and
- Page 116 and 117: to GEOGRID.TBL in the geogrid direc
- Page 118 and 119: the GRIB data were downloaded to th
- Page 120 and 121: By this point, there is generally n
- Page 122 and 123: two nests are defined to have the s
- Page 124 and 125: Note: For the WRF-NMM the variables
- Page 126 and 127: simulations may use a constant SST
- Page 128 and 129: entire simulation domain, and data
- Page 130 and 131: that lists the variables and attrib
- Page 132 and 133: http://www.ecmwf.int/products/data/
- Page 134 and 135: intermediate formats (metgrid/src/r
- Page 136 and 137: GRIB1| Level| From | To |Param| Typ
- Page 138 and 139: corresponding source grid point. Gi
- Page 140 and 141: tile_x = 1200tile_y = 1200tile_z =
- Page 142 and 143: 1. PARENT_ID : A list of MAX_DOM in
- Page 144 and 145: V -DX- H| / |DY dx DY| / |H - DX- V
- Page 146 and 147: intermediate files if ungrib is to
- Page 148 and 149: the index and data tiles for the da
- Page 150 and 151: 1. PROJECTION : A character string
- Page 152 and 153:
25. SCALE_FACTOR : A real value tha
- Page 154 and 155:
10. INTERP_MASK : The name of the f
- Page 156 and 157:
1. four_pt : Four-point bi-linear i
- Page 158 and 159:
8. search : Breadth-first search in
- Page 160 and 161:
19 Mixed Tundra20 Barren TundraTabl
- Page 162 and 163:
LU_INDEX:units = "category" ;LU_IND
- Page 164 and 165:
netcdf met_em.d01.2009-01-05_12:00:
- Page 166:
}:ISOILWATER = 14 ;:grid_id = 1 ;:p