[WRF-4.2.2] FIRE:crash: NaN detected LINE:152

[ns137]

Hello everyone, I am trying to run the Fire module of WRF 4.2.2 using real data (GFS 0.25), and I get the the following error message after running wrf.exe:

Timing for processing lateral boundary for domain 1: 0.02313 elapsed seconds
Tile Strategy is not specified. Assuming 1D-Y
WRF TILE 1 IS 1 IE 110 JS 1 JE 110
WRF NUMBER OF TILES = 1
FIRE:Time 1.500 s Average wind 0.108E+01 m/s
FIRE:Time 1.500 s Maximum wind 0.313E+01 m/s
FIRE:Time 1.500 s Fire area 0.000E+00 m^2
FIRE:Time 1.500 s Heat output 0.000E+00 W
FIRE:Time 1.500 s Max heat flux 0.000E+00 W/m^2
FIRE:Time 1.500 s Latent heat output 0.000E+00 W
FIRE:Time 1.500 s Max latent heat flux 0.000E+00 W/m^2
Timing for main: time 2024-01-22_01:00:01 on domain 1: 6.68451 elapsed seconds
FIRE:crash: NaN detected
-------------- FATAL CALLED ---------------
FATAL CALLED FROM FILE: <stdin> LINE: 152
crash: NaN detected

I verified with ncview and qgis that "wrfinput_d01" file correctly contains the Anderson vegetation classification format for the domain, and the high resolution elevation for the fire grid also appears to be correct.

Does anyone know what could be the cause, or how can I see the content of line 152?

 

[frediani]

It looks like the issue could be related to the fire ignition configuration. Some of the namelist.input variables are associated with idealized cases, for example, fire_ignition_start_x1/y1 and fire_mountain_*, but it seems you want to run a real case.

For a real case, you'd set these:
fire_ignition_start_lat1
fire_ignition_start_lon1
fire_ignition_end_lat1
fire_ignition_end_lon1
fire_ignition_radius1
fire_ignition_start_time1
fire_ignition_end_time1

Can you try that?

 

[ns137]

It looks like the issue could be related to the fire ignition configuration. Some of the namelist.input variables are associated with idealized cases, for example, fire_ignition_start_x1/y1 and fire_mountain_*, but it seems you want to run a real case.

For a real case, you'd set these:
fire_ignition_start_lat1
fire_ignition_start_lon1
fire_ignition_end_lat1
fire_ignition_end_lon1
fire_ignition_radius1
fire_ignition_start_time1
fire_ignition_end_time1

Can you try that?

Helllo frediani, thank you for answering. I just tried the parameters lat/lon but I still have the same error (crash, NaN detected, line 152). Perhaps do you have an example of a namelist.input with fire data running real.exe + wrf.exe?

 

[frediani]

Could you also check the atmospheric fields, for example, U, V, T, HGT, to see if they look OK in wrfinput_d01? In your namelist.input, the &physics section has all physics schemes set to zero, as if you're trying to run an ideal case. I don't know if this is the cause of the crash, but that's certainly an issue.
I'd suggest you turn off the fire (i.e. remove all the contents under &fire from your namelist) and run the model for the atmosphere for a short period (e.g. 30 min). Check if the main meteorological fields look realistic. If that's looking reasonable, you can add the fire again.

One namelist.input example I have handy is below. But please note that some of the settings in &physics and &domain depend on your WPS configuration. I don't know if these settings would work for you out of the box.

&time_control
run_days = 0,
run_hours = 30,
run_minutes = 0,
run_seconds = 0,
start_year = 2020,
start_month = 08,
start_day = 13,
start_hour = 18,
start_minute = 00,
start_second = 00,
end_year = 2020,
end_month = 08,
end_day = 15,
end_hour = 00,
end_minute = 00,
end_second = 00,
interval_seconds = 10800,
input_from_file = .true.,
history_interval = 30,
frames_per_outfile = 1,
restart = .false.,
restart_interval = -1,
io_form_history = 2,
io_form_restart = 2,
io_form_input = 2,
io_form_boundary = 2,
io_form_auxinput1 = 2,
io_form_auxhist2 = 2,
debug_level = 0,
/

&domains
time_step = 12,
max_dom = 1,
s_we = 1,
e_we = 210,
s_sn = 1,
e_sn = 200,
s_vert = 1,
e_vert = 45,
eta_levels = 1.00000, 0.99608, 0.99216, 0.98817,
0.98405, 0.97971, 0.97505, 0.96997,
0.96439, 0.95823, 0.95140, 0.94383,
0.93543, 0.92613, 0.91584, 0.90445,
0.89188, 0.87802, 0.86275, 0.84598,
0.82757, 0.80742, 0.78541, 0.76144,
0.73542, 0.70726, 0.67689, 0.64429,
0.60946, 0.57243, 0.53330, 0.49222,
0.44941, 0.40517, 0.35989, 0.31408,
0.26840, 0.22363, 0.18069, 0.14056,
0.10407, 0.07178, 0.04387, 0.02011,
0.00000,
p_top_requested = 20000,
num_metgrid_levels = 41,
num_metgrid_soil_levels = 9,
dx = 3000,
dy = 3000,
grid_id = 1,
parent_id = 1,
i_parent_start = 1,
j_parent_start = 1,
parent_grid_ratio = 1,
parent_time_step_ratio = 1,
feedback = 0,
smooth_option = 0,
sr_x = 30,
sr_y = 30,
/

&physics
mp_physics = 8,
ra_lw_physics = 4,
ra_sw_physics = 1,
radt = 10,
sf_sfclay_physics = 1,
sf_surface_physics = 2,
bl_pbl_physics = 5,
bl_mynn_tkeadvect = .true.,
bldt = 0,
cu_physics = 0,
cudt = 0,
isfflx = 1,
ifsnow = 1,
icloud = 1,
num_soil_layers = 4,
sf_urban_physics = 0,
maxiens = 1,
maxens = 3,
maxens2 = 3,
maxens3 = 16,
ensdim = 144,
num_land_cat = 40,
/

&dynamics
w_damping = 1
diff_opt = 1,
km_opt = 4,
diff_6th_opt = 0,
diff_6th_factor = 0.25,
damp_opt = 3,
zdamp = 5000.,
dampcoef = 0.2,
EPSSM = 0.5,
non_hydrostatic = .true.,
moist_adv_opt = 1 ,
scalar_adv_opt = 1 ,
tracer_opt = 3 ,
/

&bdy_control
spec_bdy_width = 5,
spec_zone = 1,
relax_zone = 4,
specified = .true. ,
nested = .false.,
/

&grib2
/

&namelist_quilt
/

&fire
ifire = 2,
fire_fuel_read = -1,
fire_num_ignitions = 1,
fire_ignition_ros1 = 0.05,
fire_ignition_start_lat1 = 40.609,
fire_ignition_start_lon1 = -105.879,
fire_ignition_end_lat1 = 40.609,
fire_ignition_end_lon1 = -105.879,
fire_ignition_radius1 = 250,
fire_ignition_start_time1 = 6480,
fire_ignition_end_time1 = 7000,
fire_is_real_perim = .false.,
fire_wind_height = 2.5,
fire_lsm_zcoupling = .false.,
fire_lsm_zcoupling_ref = 60.0,
/