Skip to content

V3.9.1
Compare
Choose a tag to compare
@wrf-model wrf-model released this 26 Feb 23:29
· 222 commits to master since this release
V3.9.1
d8dcb14

WRF Model Update
The WRF model tar file has been updated to Version 3.9.1 on August 17, 2017. For known problems in V3.9.1, please click here.

The WRF Pre-Processing System (WPS) has been updated to Version 3.9.1.

If you are interested in seeing how V3.9 has been tested, click here.

Improvements and Bug fixes:

Physics:

Atmospheric Radiation:

  • swint_opt: This option interpolates solar radiation between radiation steps. At sunrise the method did not use cosine zenith angle information because it only used that when it had two times to interpolate from. This results in a flat shortwave value until the second radiation time after sunrise. The new method can use zenith angle with one time, so it improves the gradient at sunrise.
  • Fixed a bug to allow Ferrier microphysics to work with GFDL radiation in WRF ARW..
  • Fixed a bug in radiation driver so that KF-CuP (cu_physics=10) clouds impact radiation.

Microphysics:

  • Thompson (mp_physics = 8,28) (Thanks to Greg Thompson of NCAR): Fixed uninitialized snow moment arrays, fixed rate_max calculation for frozen categories where density is missing, possible mass conservation voilation.

  • P3 (mp_physics = 50,51): Fixed table calculation that results in bad values, and prevented overdepletion of supersaturation during condensation calculation if long time steps are used (~ 1 min). If you're using this scheme in V3.9, get a new table from V3.9.1 (p3_lookup_table_1.dat). Also, P3 is recommended to use ONLY with RRTMG radiation. (Thanks to Hugh Morrison)

Cumulus:

  • Grell-Freitas: Fixed a bug that double counts convective ice cloud when cu_rad_feedback is on. (Thanks to Laura Fowler of NCAR)

  • G3: Fixed a compile problem due to incorrect calling of 'random_seed when using some versions of gfortran.
    Land Surface and surface layer physics:

  • NoahMP (sf_surface_physics=4): Fixed two sign errors in the initialization code for psisat. Also removes two unused arrays from the registry and initialization code and fixed FANGLE parameter calculation. (Thanks to Mike Barlage of NCAR and Gonzalo Miguez-Macho of U. Santiago de Compostela, Spain)

  • RUC LSM (sf_surface_physics=3): Improved a problem with initializing this option with GFS/NAM soil data. Before this fix, surface and boundary layer could be significant cooler and wetter when RUC LSM is initialized with GFS/NAM data. This fix improves, but does not remove the biases completely. Use with caution.(Thanks to Tanya Smirnova of NOAA)

  • PX surface layer: Changed T* calculation to use virtual potential temp at all time steps instead just the first (Thanks to Jon Pleim and Robert Gilliams of EPA).

  • Revised MM5 surface layer: Fixed calculation of PSIT and PSIQ when COARE formula is used in this scheme to be consistent with the general revision made for this scheme. Water surface only.

Dynamics:

  • Diffusion option (diff_opt = 2):
  • Fixed two mapscale factors in horizontal diffusion for scalars, which affects lat-lon projection (Thanks to Katie Lundquist of LLNL).
  • Fixed the slope term that should be outside the vertical derivative (Thanks to Pedro Jimenez of NCAR/RAL).

Initialization/Ndown:

  • Program real: Fixed map scale factor assignmeent at the norther and southern edges of a regional rotated lat/lon domain.

  • Program real: When using surface_input_source option other than 1, set urban parameters properly if they are generated from geogrid.

  • Program ndown: Fixed the bug when physics suite is used in the coarse domain run.

Software:

  • Fixed a dependency bug on a removed file (module_diag_afwa_hail.F)
Assets 2
Loading