wrong results in 10 meter wind

[simaham]

Hi everybody
I simulated wind regime over Persian Gulf with WRF for more than 30 years.
But I encounter some weird results in 2016, all the winds are southerly during first 10 months.
while in the other years everything is OK, and the model results have good accuracy in comparison with measurements and observations.
I double checked, namelist and boundary condition files of 2016, there is no difference, everything is same.
Do anyone have any idea about what happening during this year? why does it give such wrong results??
I really appreciate any help you can provide.

Sima

 

[kwerner]

Sima,
Can you let me know which version of WRF you are using (and whether it's been modified at all), and please send the namelist.input file you're using for this?
Thanks,
Kelly

 

[simaham]

Dear Kelly,

I used WRF-3.8, my namelist.input file is shown in the following:

&time_control
run_days = 29,
run_hours = 00,
run_minutes = 00,
run_seconds = 0,
start_year = 2016, 2016, 1999,
start_month = 1 , 1
start_day = 30 , 30
start_hour = 00, 00, 00,
start_minute = 00, 00, 00,
start_second = 00, 00, 00,
end_year = 2016, 2016, 1999,
end_month = 1 , 1
end_day = 31 , 31
end_hour = 06, 06, 18,
end_minute = 00, 00, 00,
end_second = 00, 00, 00,
interval_seconds = 21600
input_from_file = .true.,.true.,.true.,
history_interval = 180, 60, 60,
frames_per_outfile = 1000, 1000, 1000,
restart = .false.,
restart_interval = 5000,
io_form_history = 2
io_form_restart = 2
io_form_input = 2
io_form_boundary = 2
debug_level = 0
iofields_filename = "iofields.txt", "iofields.txt", "iofields.txt"
ignore_iofields_warning = .true.,
/

&domains
time_step = 180,
time_step_fract_num = 0,
time_step_fract_den = 1,
max_dom = 2,
e_we = 48, 103, 88,
e_sn = 36, 70, 76,
e_vert = 38, 38, 38,
p_top_requested = 5000,
num_metgrid_levels = 38,
num_metgrid_soil_levels = 4,
dx = 33353.2461, 11117.7490, 3335.32446
dy = 33353.2461, 11117.7490, 3335.32446
grid_id = 1, 2, 3,
parent_id = 1, 1, 2,
i_parent_start = 1, 7, 7,
j_parent_start = 1, 7, 7,
parent_grid_ratio = 1, 3, 3,
parent_time_step_ratio = 1, 3, 3,
feedback = 1,
smooth_option = 1
/

&physics
mp_physics = 5, 5, 1,
ra_lw_physics = 1, 1, 1,
ra_sw_physics = 1, 1, 1,
radt = 30, 30, 30,
sf_sfclay_physics = 7, 7, 1,
sf_surface_physics = 7, 7, 3,
bl_pbl_physics = 7, 7, 12,
bldt = 0, 0, 0,
cu_physics = 1, 1, 1,
cudt = 5, 5, 5,
isfflx = 1,
ifsnow = 0,
icloud = 1,
surface_input_source = 3,
num_soil_layers = 4,
num_land_cat = 24,
sf_urban_physics = 0, 0, 0,
/

&fdda
/

&dynamics
w_damping = 0,
diff_opt = 1, 1, 1,
km_opt = 4, 4, 4,
diff_6th_opt = 0, 0, 0,
diff_6th_factor = 0.18, 0.18, 0.18,
base_temp = 290.
damp_opt = 0,
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,
/

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

&grib2
/

&namelist_quilt
nio_tasks_per_group = 0,
nio_groups = 1,
/

The wierd thing is I used the same namelist file for other years, their result are normal and have good accurate,

 

[kwerner]

Hi,
What type of driving data are you using for these runs?
Can you check the driving data, not just at the initial time, but for the boundary times too (for example, you can take a look at your met_em* output) to make sure that everything looks okay? Check other fields, such as the lowest level wind in case somehow U10
and V10 are corrupt.
Thanks!

 

[simaham]

Hi
I am using Era-interim data,
I checked u10 & V10 also the lowest level of U & V Era-interim, and lowest level of met-em* files with WRF
all of them have the same trend but WRF u10 and v10 has completely different behavior in the first 10 months of 2016.
Do you think I have to control other parameters too?

I don't know how I should attach a file here. Is there any way that I could attach the picture of chart that i draw.

 

[kwerner]

Hi,
Yes, to attach files locate the blue "Options" tab below the text editing window. You'll see a tab next to that. If you click on it, it will allow you to upload attachments. This forum platform (phpBB) requires that we declare every type of attachment we will allow (e.g., .txt, .namelist, .F, etc.), so if you have a suffix that it won't allow, just add a .txt to the end of it, and then it should work. And then let me know the suffix so that I can add it to the list for future uploads. Thanks!

 

[simaham]

Hi
thanks, please look at the the attached file that shows u component of wind.
The black line indicates u10 which is extracted from WRF output file, which has completely different trend with other lines in the first 10 months of 2016.
Thank you.

 

[kwerner]

Hi,
Can you take a look at the lowest level winds to see if there is a difference? Just curious is this could potentially be a problem only with the 10m winds.

 

[simaham]

Hi,
I checked the lowest level of wind too, it is same as the 10 meter wind and gives the wrong results in the first 10 months of year 2016.
thanks

 

[kwerner]

Hi,
1) Is the plot you previously sent from a single station?
2) Can you make a wind speed plot from the model's lowest level and compare that against the driving data? In order to tell whether it is an issue with the model or driving data, you can plot the model's lowest level wind speed in a horizontal map. If the driving data is ok and the model is trying to do something else, you likely would see the model data deviate from the driving data in this plot.
3) Since you are looking at an area near a body of water, you should examine the water temperature (TSK in the model).
4) You are integrating the model for a very long time, but yet you are not using the SST update option. This is something that is always recommended for long runs. You can read more about that here: http://www2.mmm.ucar.edu/wrf/users/docs/user_guide_v4/v4.0/users_guide_chap5.html#sst_update
5) We notice that you're also using the PX LSM. You should be aware of the known limitations of that scheme. Take a look in this section of the Users' Guide and find the section that discusses this: http://www2.mmm.ucar.edu/wrf/users/docs/user_guide_v4/v4.0/users_guide_chap5.html#Phys

Kelly

 

[simaham]

Hi

I re-download the ECMWF data which I am applying as boundary condition, So the results are fine now.

Thanks

 

[kwerner]

Interesting. Well I am really glad to hear that you were able to resolve the problem! Thank you for letting us know.