Convective namelist physics

[William.Hatheway]

Good afternoon,

I have a question about which namelist options are good for convection for CONUS.

I have seen in other posts that this option is used by NCAR:

&physics
 physics_suite                       = 'CONUS'
 mp_physics                          = 28,  28,                                ! Microphysics scheme for each domain
 ra_lw_physics                       = 4, 4,                                     ! Longwave radiation scheme for each domain
 ra_sw_physics                       = 4, 4,                                      ! Shortwave radiation scheme for each domain
 radt                                = 15, 15,                                   ! Radiation time step (in minutes) for each domain
 sf_sfclay_physics                   = 4,  4,                                      ! Surface layer scheme for each domain
 sf_surface_physics                  = 4, 4,                                     ! Land surface model for each domain
 sf_urban_physics                    = 0, 1,                                      ! Urban physics scheme for each domain
 bl_pbl_physics                      = 4, 4,                                      ! Planetary boundary layer (PBL) scheme for each domain
 bldt                                = 0, 0,                                      ! Time step for PBL schemes (in minutes) for each domain
 cu_physics                          = 11, 0,                                     ! Cumulus parameterization scheme for each domain
 cudt                                = 5,  5,                                     ! Time step for cumulus schemes (in minutes) for each domain
 isfflx                              = 1,                                       ! Surface flux computation option
 ifsnow                              = 1,                                       ! Snow-cover effect on surface albedo option
 icloud                              = 1,                                       ! Cloud water/ice effect on radiation option
 surface_input_source                = 1,                                       ! Source for surface input data
 num_soil_layers                     = 4,                                       ! Number of soil layers in the model
 num_land_cat                        = 21,                                      ! Number of land-use categories

/

But the namelist CONUS physics package uses this:

&physics
 physics_suite                       = 'CONUS'
 mp_physics                          = 8,  8,                                ! Microphysics scheme for each domain
 ra_lw_physics                       = 4, 4,                                     ! Longwave radiation scheme for each domain
 ra_sw_physics                       = 4, 4,                                      ! Shortwave radiation scheme for each domain
 radt                                = 15, 15,                                   ! Radiation time step (in minutes) for each domain
 sf_sfclay_physics                   = 2,  2,                                      ! Surface layer scheme for each domain
 sf_surface_physics                  = 4, 4,                                     ! Land surface model for each domain
 sf_urban_physics                    = 0, 0,                                      ! Urban physics scheme for each domain
 bl_pbl_physics                      = 2, 2,                                      ! Planetary boundary layer (PBL) scheme for each domain
 bldt                                = 0, 0,                                      ! Time step for PBL schemes (in minutes) for each domain
 cu_physics                          = 6, 0,                                     ! Cumulus parameterization scheme for each domain
 cudt                                = 0,  0,                                     ! Time step for cumulus schemes (in minutes) for each domain
 surface_input_source                = 1,                                       ! Source for surface input data
 num_soil_layers                     = 4,                                       ! Number of soil layers in the model
 num_land_cat                        = 21,                                      ! Number of land-use categories

/

Are both interchangeable or is one better than the other?

 

[Ming Chen]

No, I don't think one suite of options is better than the other suite. This is because WRF performance is always case-dependent. Thus which suite is better cab only be determined based on results.