Source code

Latest version

Version : 5.5.1

Old versions


5.5.1 ( bug fix from 5.5.0 )
  • Fixed the mkinit for the baroclinic wave test case
  • Fixed a compile error for some version of GNU compiler
  • Change default setting for MPI for FUJITSU compiler
  • Updated the user guide
5.5.0 ( minor updates from 5.4.5 )
  • Unified NetCDF file input module for the real input
  • Default use of domain center as the standard latitude for Mercator and equidistant cylindrical projections
  • Eliminated need to give the latlon catalog file
  • Eliminated need to give time information for boundary data
  • Eliminated need to give parent domain process information for offline nesting
  • Updated the namelist file and conversion script for the latest FNL data
  • Changed format of timer information in LOG file
  • Support newer version of GNU compiler
  • Addition and removal of several sysdef files
  • Added several options to the Tomita08 scheme
  • Added several options to the SN14 scheme
  • Added several options to the MYNN scheme
  • 2D parameters are available in the Kusaka01 scheme
  • User-defined calendar can be used
  • Added ability to output time min/max data to history file
  • Allowed relative paths to be set for bulk job
  • Tutorial simulations now support GPU computers
  • Supported one-sided MPI communication
  • Improved iterations in the land and urban models
  • Improved numerical stability for the MYNN level 3 scheme
  • Improved handling of thermodynamical quantities in the KF scheme
  • Improved momentum flux calculation in the Kusaka01 scheme
  • Reduce memory usage in the real input
  • Optimization for the Fugaku supercomputer
  • Optimization of the SUZUKI10 scheme
  • Fixed some bugs in SNO
  • Fixed many other bugs
5.4.5 ( bug fix from 5.4.4 )
  • Fix precipitation calculation for energy conservation
  • Fix the grid nudging for REAL-type boundary
  • Fix a stalling problem at single file output
  • Fix for the newer versions of GNU compiler
  • Update sysdep files for FUGAKU
5.4.4 ( bug fix from 5.4.3 )
  • Fix a bug in the SN14 scheme (thanks to @slayoo)
5.4.3 ( bug fix from 5.4.2 )
  • Fix wrong OpenMP directives
  • Fix out-of-array reference in the vertical interpolation
  • Fix incorrect calculation of the vertical divergence
  • Fix the horizontal interpolation of zonally cyclic data
  • Update the sysdef files for FUGAKU
5.4.2 ( bug fix from 5.4.1 )
  • Fix a bug in the SNO to output topo and lsmask from pp files.
  • Fix bugs for single precision calculation
  • Fix wrong openmp directives
  • Fix wrong number density of the hydrometeors in the real input
  • Change the calculation of the optical parameters in the mstrnX to avoid extraporation
5.4.1 ( bug fix from 5.4.0 )
  • Fix incorrect changes in surface temperature when ATMOS_PHY_SF_TYPE is CONST.
  • Fix a problem with different results in restart runs
  • Fix a bug in SNO
  • Fix a bug in the UD7 and CD8 flux schemes
  • Fix undefined references
  • Change some options for the FUGAKU supercomputer
5.4.0 ( minor updates from 5.3.6 )
  • Change the configuration parameters of the KUSAKA model
  • Change the configuration parameters of the KF scheme
  • Change the configuration parameters for the Coriolis parameter
  • Change the configuration parameters for topography
  • Change the configuration parameters for GrADS input
  • Change the configuration parameter for the depth of the slab ocean
  • Add the SATO2019 lightning scheme (experimental)
  • Add a grid and spectral nudging
  • Add a seven-step RK scheme
  • Add sysdep files for LinuxARM, FUGAKU, OBCX, and NIESSX computers
  • Add the bi-linear interpolation to the real input and online nesting
  • Support freezing of soil moisture
  • Enable two-dimensional calculation
  • Enable to output the Obukhov length and scales to history output
  • Enable to use of all even orders for the numerical filter
  • Enable to interpolate to the pressure coordinate in SNO
  • Enable to use a different map projection than the parent domain for the SCALE offline nesting
  • Change ocean and land calculations for better conservation of the total energy
  • Change calculation of the hydrostatic pressure
  • Change calculation of initial state in the real input
  • Fix bug in the SUZUKI10 scheme
  • Fix some bugs in SNO
  • Fix some bugs in the KF scheme
  • Fix some bugs in the NYNN scheme
5.3.6 ( bugfix from 5.3.5 )
  • Fix wrong calculation of T2 and Q2
  • Fix wrong array index in the precipitation scheme
  • Fix wrong a coefficient of the universal function in the bulkflux scheme
  • Fix some bugs in the MYNN scheme
  • Fix MPI error in SNO
  • Fix negative fixer to conserve moist inernal energy
  • Fix for positive definite in the KF scheme
  • Fix potential floating overflow in the TOMITA08 scheme
  • Fix bug in storing history variables in the TOMITA08 scheme
  • Fix wrong configuration for the WK1982 test case
  • Change calculation of MSLP
  • Change to avoid floating point exception with Intel compiler
5.3.5 ( bugfix from 5.3.4 )
  • Fix wrong local time in the urban model
  • Fix wrong land use factor
5.3.4 ( bugfix from 5.3.3 )
  • Fix unexpected decrease of the soil temperature with snow precipitation
  • Fix unexpected decrease of the urban temperature with evaporation
  • Fix wrong the TKE production at the bottom layer in the MYNN scheme
  • Fix some problems with the intel compiler
  • Fix some problems with the recent FNL data in the real init process
  • Fix the wrong wind direction in the tutorial real case
5.3.3 ( bugfix from 5.3.2 )
  • Fix the breakup of the rain in the SN14 scheme
5.3.2 ( bugfix from 5.3.1 )
  • Fix wrong layer depth in the KF scheme
  • Fix some bugs related to the map factor
  • Fix some problems in the real init with a certain condition
5.3.1 ( bugfix from 5.3.0 )
  • Fix the vapor flux at the surface
  • Fix wrong heat capacity in the KF scheme
  • Fix wrong roughness length in the B91W01
  • Fix potential segmentation error in the real init
5.3.0 ( minor updates from 5.2.6 )
  • Change the ocean grid to have the vertical dimension
  • Change several namelist parameter names; provide a configure converter
  • Change the smoothing filter for the topography
  • Change internal structure
  • Change the radiation flux category from short and long waves to IR, near-IR and visible.
  • Change the category of the planetary boundary layer scheme from TB to BL
  • Increase grid size of the variables at the staggered point in the netcdf file.
  • Add several attributes into the netcdf file.
  • Add some examples for analysis using SCALE library
  • Add a new post-process tool SNO
  • Add several prognostic variables to history output
  • Add the RN222 scheme
  • Add ideal 360- and 365-day calendars
  • Add a semi-Lagrangian precipitation scheme
  • Add a sea ice scheme
  • Enable to nudge the ocean and land variables
  • Enable to use rotated domain
  • Enable to control solarins for ideal experiments
  • Enable to set the Coriolis parameter for ideal experiments
  • Enable to split the history output into multiple files for the time dimension
  • Enable to split external input files for the time dimension
  • Change for the better initial states especially near the surface
  • Reduce memory usage in the initial process
  • Change error message to more user friendly
  • Fix wrong check of domain size for the nest setting
  • Fix a bugs in the SMG scheme related to the terrain-following coordinate
  • Many minor bug fixes
5.2.6 ( bugfix from 5.2.5 )
  • Fix wrong flux of tracers at the lateral boundaries
5.2.4 ( bugfix from 5.2.5 )
  • Fix the surface precipitation flux to include that calculated by the cumulus parameterization
  • Add some parameters of NAMELIST groups to the list of the reference which is missed in the list
  • Fix a bug in the initialization process of SCALE-GM
  • Fix a bug in unit for the time in the ico2ll
  • Change default value of "Z_LEV_TYPE" of the net2g from "plev" to "original"
5.2.4 ( bugfix from 5.2.3 )
  • Fix bugs in calculation of the cloud fraction and effective radius in the Suzuki10 scheme
  • Fix a bug in checking domain size
  • Fix wrong axes in file: LCDZ, UCDZ
  • Fix an error in generating reference manual
5.2.3 ( bugfix from 5.2.2 )
  • Fix the nuding at the lateral boundary region with the SN14 and Suziki10 schemes
  • Fix the CP scheme in the case with the SN14 scheme
  • Fix the online nesting in the case using the Tomita08 and SN14 scheme in the parent and daughter domains, respectively
5.2.2 ( bugfix from 5.2.1 )
  • Fix wrong loop range in the SUZUKI10 bin microphysics scheme
  • Fix wrong soil water content in the init process with the WRF file
  • Fix wrong density in the init process with the SCALE file
  • Fix the grib2grads converter for MacOS
5.2.1 ( bugfix from 5.2.0 )
  • Change the netcdf checker for Cent OS and Fedora
  • Fix the friction velocity in the MYNN scheme
  • Fix a sign of the convective velocity scale in the B91W01 scheme
  • Fix unit of the time dimension in the history file
  • Fix spurious land fraction
  • Fix wrong soil water at the bottom land layer
5.2.1 ( minor updates from 5.2.0 )
  • Add prefix of 'SCALE_' to all the environmental variable names
  • Change namelist parameters for the nesting
  • Change namelist parameters for the horizontal eddy viscosity
  • Change name of the TKEs in the turbulent schemes for outputs
  • Split the Rayleigh damping at upper layer from nudging in the buffer region
  • Enable to give external radiation forcing
  • Add Thick-SLAB land surface scheme
  • Add new options for the Tomita08 scheme
  • Enable to set the number of grids in global domain
  • Enable to set buffer size with the number of grids
  • Enable to output complete namelist parameter sets to a file separately from the log file
  • Support single file I/O with PnetCDF (requires PnetCDF >= 1.7.0)
  • Enable to output history data with pressure and height coordinates
  • Add options for restart file name
  • Support converter for FNL data with grib2 format for the real init
  • Support different grid points for atmosphere, land and ocean data for the real init
  • Provide English User Guide
  • Update table for radiational data
  • Improve spatial interpolation
  • Implore numerical stability for steep terrains
  • Fix bugs in the partial condensation in the MYNN scheme
  • Fix several bugs in the KF scheme
  • Fix wrong Prandtle number for stable case in the B91W01 scheme
  • Many minor bug fixes
5.1.5 ( bugfix from 5.1.4 )
  • Fix wrong land fraction when GLCCv2 data is used
5.1.4 ( bugfix from 5.1.3 )
  • Fix wrong cloud fraction of the Tomita08 scheme
5.1.3 ( bugfix from 5.1.2 )
  • Fix wrong gamma function
5.1.2 ( bugfix from 5.1.0 )
  • Fix wrong metric for the terrain-following coordinate
  • Fix wrong tendency of the density
  • Fix wrong sign of the metric terms
  • Fix invalid value of some monitor variables
5.1.0 ( minor updates from 5.0.1 )
  • Change the tutorial test cases
  • Update documents
  • Add the Kain and Fritsch (1990) cumulus parameterization (experimental)
  • Provide the script to prepare the configurations for real atmospheric experiments
  • Implove netCDF I/O handling
  • Fix bug in the Suzuki10 scheme
  • Fix treatment the boundary data for restart experiment
  • Fix handling for cyclic outer data
5.0.1 ( bugfix from 5.0.0 )
  • Fix wrong dependency for making SCALE-GM
5.0.0 ( major updates from 0.3.3 )
  • change name SCALE-LES to SCALE-RM (also name of binary)
  • change time format of restart file name (yyyymmdd-hhmmss.mmm)
  • Enable to select schemes for advection flux (CD2, UD3, UD3Koren1993, CD4, UD5, CD6; default is changed)
  • Enable to select schemes for time step (RK3, RK3WS2002, RK4; default is changed)
  • Enable to real temporally varied sea surface temperature data
  • Enable to integrate with tendencies of physical processes without advection
  • Enable to output clear-sky radiation flux
  • Add the Deardroff (1980) scheme for SGS turbulence
  • Add the Kajino (2013) scheme for aerosol microphisics (experimental)
  • Enable to use multiple data sources for topography and landuse
  • Enable to put barrier at call of the rap time routine
  • change algorighm of local filter in smoothing topography
  • Fix wrong calculation of internal energy
  • Fix negative fixer in the Tomita08 scheme
0.3.3 ( bugfix from 0.3.1 )
  • Fix a bug in the divergence dumping
  • Fix possible zero division in the MSTRN-X scheme
0.3.1 ( bugfix from 0.3.0 )
  • Fix a bug in the Suzuki10 scheme used with the MSTRN-X scheme
  • Fix wrong liquid water content in calculation of the hydrostatic profile in the real init
0.3.0 ( minor updates from 0.2.6 )
  • implove the init program for real case
  • enable to change parameters of the Tomita08 scheme
  • enable to output some statistics (CAPE, CIN, PBL height, etc)
  • confirm the single precision calculation
  • add a CFL checker
  • implove computational performace
  • many bugfixes

Known issues

  • SCALE-5.3.x cannot be compiled with the intel compiler version or later due to a bug in the compiler. Download Makefile and replace scalelib/src/Makefile with the file to compile by the intel compiler.


  • Parameter input files MSTRNX parameter files (scale/scale-rm/test/data/rad) were updated in version 5.2. It is recommended to use the new parameter files provided in the source code, when using the version 5.2 or later.


The documents for SCALE are available at SCALE Document page


Topo and landuse data

  • scale_database.tar.gz
    • update: 2015/07/22
    • size: 759MB (795325305 B)
    • MD5: f63cff8a209e73e8ceac71e33becbd60

Horizontal grid and icosahedral grid for SCALE-GM

latlon mapping table for SCALE-GM