WRF Model Update
WRF model tar file has been updated to Version
3.1 on April 9, 2009. For WRF Version 3.1.1 updates (7/31/09), please click here.
WRF Pre-Processing System (WPS) has been updated to Version 3.1.
Note: The V3.1 WRF can use V3.0 wrfinput and wrfbdy files as long as you don't use the gravity drag option which requires new data from geogrid, and add a new namelist variable, use_baseparam_fr_nml in &dynamics. Rerunning the new programs is recommended. New options have not been tested extensively. Please use with caution.
Also see 'Known
Problems' in this release. If you are interested in seeing
how V3 has been tested, click here.
In this new release, the following new features and updates are provided:
New in Version 3.1:
- Thompson microphysics (from Greg Thompson of NCAR/RAL)
- WRF Double Moment schemes for 5 and 6-class microphysics (from K. S. Lim and S. Y. Hong of Yonsei University, South Korea)
- Quasi-normal scale elimination (QNSE) PBL and surface layer schemes (from B. Galperin [U. S. Florida) and S. Sukoriansky [Ben Gurion Univ of Israel])
- Mellor-Yamada Nakanishi Niino (MYNN) level 2.5 and level 3 PBL and MYNN surface schemes (from M. Pagowski of NOAA/GSD)
- Bougeault-Lacarrere PBL (BouLac, from A. Martilli of CIEMAT, Spain)
- Building Environment Parameterization, a multi-layer urban canopy model (from A. Martilli (CIEMAT, Spain), S. Grossman-Clarke (Arizona State U), M. Tewari of NCAR). BEP currently only works with MYJ and BouLac (new in 3.1) PBL options (read more)
- MODIS 20 category landuse data for Noah LSM
- RRTMG long- and short-wave radiation schemes (from M. Iacono and E. Mlawer of AER) (see paper by Iacono et al.)
- Fractional sea-ice and modification for snow/ice LSM physics (from BPRC of OSU)
- Deep soil temperature update and diurnal variation of SST; bucket option for precipitation (from R. Leung and X.-B. Zeng) and energy flux accumulation for regional climiate application
- Gravity wave drag parameterizaton (from S. Y. Hong and H. Y. Shin of YSU)
- Monotonic transport scheme
- Gridded surface analysis nudging (from A. J. Deng and D. R. Stauffer of PSU)
- Spectral nudging (using gridded analysis) (from G. Miguez-Macho (U Santiago de Compostela, Spain)
- New idealized case: single column model (from J. Hacker of NCAR/RAL)
- Introduce a new parameter to define base state temperature in stratosphere. This allows the model top to be extended to 5 mb (iso_temp)
- Option to start moving nest at some specified time (time_to_move)
- Capability to interpolate soil data from any external source to what is required in the model
- SKINTEMP is no longer a required field, it may be substituted by diurnally averaged surface temp (TAVGSFC computed using WPS utility program avg_tsfc.exe) by using namelist variable use_tavg_for_tsk.
- A simple TC bogus option using tc.exe. Works for isobaric data after WPS. Run after WPS and before real.
Improvements and Bug fixes:
- Ferrier (update to NCEP operational version)
PBL and surface layer physics:
- MYJ PBL and MYJ surface layer physics (update to NCEP operational version)
Land surface and urban physics:
- Noah LSM: albedo, z0 and emissivity calculation dependent on vegetation fraction; no longer use data from LANDUSE.TBL; snow albedo change
- UCM fixes; table name changed to URBPARM.TBL
- PX LSM: albedo change and fixes, nesting fix.
- CAM: add CO2 variation; add no-leap year option; fix CO2 for shortwave; fix aerosols for longwave; add volcanic aerosol initialization for longwave-only CAM.
- KF: fixed a problem introduced when adaptive time step was implemented.
- old KF: fixed a problem using -O3 -qhot optoin on IBM.
- Fixed wrong GMT info in wrfout file when grid analysis nudging (wrffdda) is used.
- Bug fixes for observation nudging
- namelist variable pd_moist, pd_tke, pd_scalar and pd_chem (logical switch) are now moist_adv_opt, tke_adv_opt, scalar_adv_opt and chem_adv_opt (real number of 1). This is to accomodate a second option in this category, the monotomic option (2).
- tke bc3d correction for restart
- minor fixes for polar fft
- vertical motion fix at nest moving time (this does affect results)
- unify model constants for idealized case initialization.
- Fixed for occasional failure of real when surface is below sea level.
- Option to compute qvapor with respect to water or ice in real.
- Fixed for computing dry atmospheric pressure when geopotential height is not available in real. Geopotential height will first be estimated now rather than ignored.
- Base state parameters (base_temp, base_pres, base_lapse, iso_temp) are now written to wrfinput file for model to use. Avoid using different values in real and wrf. If V3.1 real is not used, it will require the use of a new namelist, use_baseparm_fr_nml, to tell wrf you are using the namelist values.
- Fixed a problem associated with nested terrain adjustment and cycled model runs.
- Fixed for occasional failure of adaptive time stepping, mostly at restart time.
- Fixed errors associated with nest ratios 4 and 5.
- Wind rotation fix in time series output.
- ucmcall is changed to sf_urban_physics
- pd_moist/tke/scalar/chem are changed to moist/tke/scalar/chem_adv_opt
- ESMF 3.1
- Relax restrictions on number of processors decomposing a domain (so-called “over-decomposition”)
- Extra dimensions to 4-D tracer arrays
- Improve compilation
a. Break up large routines, e.g. solve_em
b. Use ONLY clause
c. OpenMP compilation improvements
d. Automatic test for NetCDF large file support
- IBM default optimization turned up
-DNATIVE_MASSV with -lmass and -lmassv
- Registry changes
a. Package dependent allocation of state variables
b. Multi-character dimspecs