wrf.exe ends without CFL error (instability)

[delmoral]

Hi!
I am new in WRF and I've been struggling for two weeks with my domains. I am running WRF 4.0.3 over the Iberian Peninsula, specially the NE corner (Pyrenees inside domain). I am running three nested domains (12-3-1km) and neither the 3 km nor the 1 km is ending the simulation. At the beginning I was getting CFL errors but now, I am not getting any error.

I am using all the options available and I have also smoothed my topography in the GEOGRID.TBL file (option=1-2-1 and 4 passes). I am also using static data with not so high resolution.


Here is my namelist.input:
&time_control
run_days = 0,
run_hours = 6,
run_minutes = 0,
run_seconds = 0,
start_year = 2018, 2018, 2018,
start_month = 01, 01, 01,
start_day = 06, 06, 06,
start_hour = 21, 21, 21,
end_year = 2018, 2018, 2018,
end_month = 01, 01, 01,
end_day = 07, 07, 07,
end_hour = 03, 03, 03,
interval_seconds = 10800,
input_from_file = .true.,.true.,.true.,
history_interval = 30, 30, 10,
frames_per_outfile = 1, 1, 1,
restart = .false.,
restart_interval = 30,
io_form_history = 2,
io_form_restart = 2,
io_form_input = 2,
io_form_boundary = 2,
debug_level = 9999,
/

&domains
time_step = 60,
time_step_fract_num = 0,
time_step_fract_den = 1,
max_dom = 2,
e_we = 80, 196, 250,
e_sn = 80, 196, 181,
e_vert = 64, 64, 64,
p_top_requested = 15000,
num_metgrid_levels = 32,
num_metgrid_soil_levels = 4,
dx = 15000, 3000, 1000,
dy = 15000, 3000, 1000,
grid_id = 1, 2, 3,
parent_id = 0, 1, 2,
i_parent_start = 1, 30, 50,
j_parent_start = 1, 20, 50,
parent_grid_ratio = 1, 5, 3,
parent_time_step_ratio = 1, 5, 3,
feedback = 0,
smooth_option = 0,
smooth_cg_topo = .true.,
target_cfl = 1.05,1.05,
use_adaptive_time_step = .false.,
target_hcfl = 1.05,1.05,
starting_time_step = 75, 15, 5,
max_step_increase_pct = 100,100,100,
max_time_step = 200,40,5,
min_time_step = 30,6,2,
adaptation_domain = 1,
sfcp_to_sfcp = .true.,
/

&physics
mp_physics = 8, 8, 8,
cu_physics = 16, 0, 0,
ra_lw_physics = 5, 5, 5,
ra_sw_physics = 5, 5, 5,
bl_pbl_physics = 8, 8, 8,
sf_sfclay_physics = 1, 1, 1,
sf_surface_physics = 1, 1, 1,
radt = 5, 5, 5,
bldt = 0, 0, 0,
cudt = 0, 0, 0,
icloud = 1,
num_land_cat = 21,
sf_urban_physics = 0, 0, 0,
num_soil_layers = 5,
/

&fdda
/

&dynamics
w_damping = 1,
diff_opt = 1, 1, 1,
km_opt = 4, 4, 4,
diff_6th_opt = 0, 0, 0,
diff_6th_factor = 0.12, 0.12, 0.12,
base_temp = 290.
damp_opt = 3,
zdamp = 5000., 5000., 5000.,
dampcoef = 0.2, 0.2, 0.2,
khdif = 0, 0, 0,
kvdif = 0, 0, 0,
non_hydrostatic = .true., .true., .true.,
moist_adv_opt = 1, 1, 1,
scalar_adv_opt = 1, 1, 1,
/ gwd_opt = 1,
epssm = 1,
/

&bdy_control
spec_bdy_width = 5,
spec_zone = 1,
specified = .true., .false., .false.,
nested = .false., .true., .true.,
/

&grib2
/

&namelist_quilt
nio_tasks_per_group = 0,
nio_groups = 1,
/


Any help is more than welcome.
Thank you so much.

 

[meteoadriatic]

Hello,

I'm sure you are still getting CFL vilations but probably WRF segfaults before it prints them out into log. And this is almost guaranteed one of problems:

max_step_increase_pct = 100,100,100,

You're directing adaptive time step to increase 100 percent if cfl is below target value. That is huge increase. Please try default values first (note, for parent domain default value is just 5 percent!).

Ivan

 

[Ming Chen]

Your namelist.input looks fine, except that you can set radt=15, 15 for the case.

Can you try to run a single domain, and see how it works?

Please also try the following options:
ra_lw_physics = 4, 4
ra_sw_physics = 4, 4
bl_pbl_physics =1, 1
cu_physics=1, 0

These tests will help us narrow down the problem.