Page 1 of 1

MYNN2.5 with CFL errors at the edge of nested domains

Posted: Wed Mar 24, 2021 12:50 am
by bartbrashers
I have been testing the MYNN2.5 PBL scheme in stable conditions (wintertime in Alaska) for several projects, some successfully, some not.

Using WRF-4.2.2, with nested 12, 4, and 1.33 km domains, I see lots of CFL errors in the "margins", but never anywhere else.

My d02 is 202 x 202 grid points, and I 'm using 39 vertical levels. The CFL errors occur consistently in the 2nd grid cell from the edge of the domain and consistently at the next-to-top level:

Code: Select all

# grep -i cfl ????-??-??/wrf.rsl.error.00*
2019-12-05/wrf.rsl.error.0028:d02 2019-12-04_13:40:00             1  points exceeded cfl=2 in domain d02 at time 2019-12-04_13:40:00 hours
2019-12-05/wrf.rsl.error.0028:d02 2019-12-04_13:40:00  MAX AT i,j,k:             2          181           38  vert_cfl,w,d(eta)=    2.203470       -18.77660       1.7999999E-02
2019-12-05/wrf.rsl.error.0028:d02 2019-12-04_15:27:24             2  points exceeded cfl=2 in domain d02 at time 2019-12-04_15:27:24 hours
2019-12-05/wrf.rsl.error.0028:d02 2019-12-04_15:27:24  MAX AT i,j,k:             2          199           38  vert_cfl,w,d(eta)=    2.977944       -40.41059       1.7999999E-02
2019-12-05/wrf.rsl.error.0031:d02 2019-12-04_14:04:36             1  points exceeded cfl=2 in domain d02 at time 2019-12-04_14:04:36 hours
2019-12-05/wrf.rsl.error.0031:d02 2019-12-04_14:04:36  MAX AT i,j,k:           199          200           38  vert_cfl,w,d(eta)=    2.241721       -19.70888       1.7999999E-02
2019-12-05/wrf.rsl.error.0031:d02 2019-12-04_14:53:48             1  points exceeded cfl=2 in domain d02 at time 2019-12-04_14:53:48 hours
2019-12-05/wrf.rsl.error.0031:d02 2019-12-04_14:53:48  MAX AT i,j,k:           196          200           38  vert_cfl,w,d(eta)=    3.073648       -35.45598       1.7999999E-02
2019-12-05/wrf.rsl.error.0031:d02 2019-12-04_17:54:24             7  points exceeded cfl=2 in domain d02 at time 2019-12-04_17:54:24 hours
2019-12-05/wrf.rsl.error.0031:d02 2019-12-04_17:54:24  MAX AT i,j,k:           200          199           38  vert_cfl,w,d(eta)=    5.358111       -19.34883       1.7999999E-02
2019-12-20/wrf.rsl.error.0024:d03 2019-12-22_20:02:45             1  points exceeded cfl=2 in domain d03 at time 2019-12-22_20:02:45 hours
2019-12-20/wrf.rsl.error.0024:d03 2019-12-22_20:02:45  MAX AT i,j,k:             2          169           13  vert_cfl,w,d(eta)=    2.059243       -13.70688       1.5999973E-02
2019-12-30/wrf.rsl.error.0027:d02 2019-12-29_18:25:00             5  points exceeded cfl=2 in domain d02 at time 2019-12-29_18:25:00 hours
2019-12-30/wrf.rsl.error.0027:d02 2019-12-29_18:25:00  MAX AT i,j,k:            87          200           38  vert_cfl,w,d(eta)=    2.956032       -36.84430       1.7999999E-02
2019-12-30/wrf.rsl.error.0031:d02 2019-12-29_13:29:12             2  points exceeded cfl=2 in domain d02 at time 2019-12-29_13:29:12 hours
2019-12-30/wrf.rsl.error.0031:d02 2019-12-29_13:29:12  MAX AT i,j,k:           197          200           38  vert_cfl,w,d(eta)=    2.357880       -21.55417       1.7999999E-02
2019-12-30/wrf.rsl.error.0031:d02 2019-12-29_16:22:00             2  points exceeded cfl=2 in domain d02 at time 2019-12-29_16:22:00 hours
2019-12-30/wrf.rsl.error.0031:d02 2019-12-29_16:22:00  MAX AT i,j,k:           199          200           38  vert_cfl,w,d(eta)=    2.172655       -19.27447       1.7999999E-02
2019-12-30/wrf.rsl.error.0031:d02 2019-12-29_17:17:48             1  points exceeded cfl=2 in domain d02 at time 2019-12-29_17:17:48 hours
2019-12-30/wrf.rsl.error.0031:d02 2019-12-29_17:17:48  MAX AT i,j,k:           200          197           38  vert_cfl,w,d(eta)=    2.382340       -17.79342       1.7999999E-02
2019-12-30/wrf.rsl.error.0031:d02 2019-12-29_18:52:36             1  points exceeded cfl=2 in domain d02 at time 2019-12-29_18:52:36 hours
2019-12-30/wrf.rsl.error.0031:d02 2019-12-29_18:52:36  MAX AT i,j,k:           198          200           38  vert_cfl,w,d(eta)=    2.026072       -16.73941       1.7999999E-02
Aren't CFL errors suppressed in points near a domain boundary by the parent domain applying boundary conditions on the child domain? And why would a CFL error happen at the TOP of the domain?

Some relevant chunks of my namelist.input:

Code: Select all

 max_dom                             = 3,
 p_top_requested                     = 5000,
 e_vert                              = 39,      39,      39,      
 eta_levels                          = 1.0000,  0.9995,  0.9990,  0.9984,  0.99705,
                                       0.99415, 0.99155, 0.9860,  0.9780,  0.9660,
                                       0.9500,  0.9340,  0.9180,  0.9020,  0.8860,
                                       0.8660,  0.8420,  0.8140,  0.7800,  0.7400,
                                       0.6940,  0.6480,  0.6020,  0.5560,  0.5100,
                                       0.4640,  0.4180,  0.3720,  0.3260,  0.2820,
                                       0.2400,  0.2000,  0.1630,  0.1280,  0.0960,
                                       0.0660,  0.0400,  0.0180,  0.0000,

 use_adaptive_time_step              = .true.,
 step_to_output_time                 = .true.,
 adaptation_domain                   = 3,
 target_cfl                          =  0.6,   0.6,   0.6,   
 target_hcfl                         =  0.6,   0.6,   0.6, 
 max_step_increase_pct               =    3,    51,    51,    
  starting_time_step                  =    6,     2,     1,    
 max_time_step                       =   36,    12,     4,   
 min_time_step                       =    1,     1,     1,   
 min_time_step_den                   =    1,    10,    10, 
 
 &physics
 mp_physics                          = 10,     10,     10,   
 ra_lw_physics                       = 4,     4,     4,  
 ra_sw_physics                       = 4,     4,     4,   
 radt                                = 20,   20,    20,    
 sf_surface_physics                  = 2,     2,     2,    
 sf_sfclay_physics                   = 5,     5,     5,    
 bl_pbl_physics                      = 5,     5,     5,     
 bldt                                = 0,     0,     0,   
 cu_physics                          = 1,   0,    0, 
 cudt                                = 0,     0,     0,   
 kfeta_trigger                       = 2,
 
Any insights or suggestions would be greatly appreciated!

Bart

Re: MYNN2.5 with CFL errors at the edge of nested domains

Posted: Tue Mar 30, 2021 11:30 pm
by kwerner
Hi Bart,
Can you attach your full namelist so that I can take a look? Thanks!