WRF - extreme wind speeds in mountains

Z

Z Grajek

New member
Hi, I'm struggling with simulating extreme wind in a mountainous terrain. Below is a graph of wind speed at the mountain station vs. the model. The model couldn't cope with the speed drops towards the end of the simulation. These are results from a 300-meter domain. I've attached the nameless.input I used. Does anyone have any advice on how to manage wind speeds in the mountains? Any tips? Best regards Zosia
1783428402574.png

&domains
time_step = 4,

max_dom = 3,

e_we = 310, 265, 491,
e_sn = 330, 331, 721,
i_parent_start = 1, 115, 100,
j_parent_start = 1, 130, 114,
e_vert = 70, 70, 70,
auto_levels_opt = 2,
dzbot = 10.0,
dzstretch_s = 1.1,
dzstretch_u = 1.1,

p_top_requested = 5000,
num_metgrid_levels = 38,
num_metgrid_soil_levels = 4,
dx = 4500,
dy = 4500,
grid_id = 1, 2, 3,
parent_id = 1, 1, 2,
parent_grid_ratio = 1, 3, 5,
parent_time_step_ratio = 1, 3, 5,
feedback = 1,
smooth_option = 1,
numtiles = 4,
smooth_cg_topo = .false.,

nproc_x = 16
nproc_y = 16
/

&physics
mp_physics = 8, 8, 8,
ra_lw_physics = 4, 4, 4,
ra_sw_physics = 4, 4, 4,
radt = 3, 3, 3,
sf_sfclay_physics = 5, 5, 5,
sf_surface_physics = 2, 2, 2,
num_soil_layers = 4,
bl_pbl_physics = 5, 5, 5,
bldt = 0, 0, 0,
cu_physics = 0, 0, 0,
cudt = 0, 0, 0,
isfflx = 1,
ifsnow = 1,
icloud = 1,
surface_input_source = 3,
num_land_cat = 21,
sf_urban_physics = 0, 0, 0,
bl_mynn_edmf = 1, 1, 1,
bl_mynn_edmf_tke = 1, 1, 1,
bl_mynn_tkeadvect = .true., .true., .true.,
/

&fdda
/

&dynamics
hybrid_opt = 2,
non_hydrostatic = .true., .true., .true.,
w_damping = 1,
damp_opt = 3,
zdamp = 5000., 5000., 5000.,
dampcoef = 0.2, 0.2, 0.2,
diff_opt = 2, 2, 2,
km_opt = 4, 4, 4,
diff_6th_opt = 2, 2, 2,
diff_6th_factor = 0.02,
epssm = 0.6, 0.6, 0.6,
base_temp = 290.,
moist_adv_opt = 1, 1, 1,
scalar_adv_opt = 1, 1, 1,
gwd_opt = 0, 0, 0,
/
 
Your namelist.input looks fine to me except that:

(1) time step can be larger. For dx=4.5km, max time_step can be 27s
(2) it may be better to run this case with the SMS-3DTKE scale-adaptive LES/PBL scheme
(3) Please turn on the option topo_wind, which may be helpful to yield better surface winds

Hope this is helpful. Surface wind simulation is always a challenging issue in WRF studies. I know that some people develop post-process tools to correct bias in their simulations.

By the way, how did you match model results with your station observations? Did you interpolate model data to station sites? if so, how did you do ?

Please let me know if you have more questions.
 
Hi Ivan,

You are right that topo_wind only works with YSU PBL. Sorry that I forgot this prerequisite for using this option.

Ming
 
1784740032213.png
Hi, I'm using the grid closest to the station. I'm using ASTER 30m data with smooth_passes=3, so the station elevation difference is 1990, and the model is 1908, which seems like a good value to me. The graph shows the mountain station on the left, and the leeward side of the mountains on the right. Is it possible to get something like a pink line on the graph at these high speeds? Because the model clearly doesn't detect when to reduce the speed. I'm also concerned about the strange speed peaks at the station on the leeward side.
 
@ Z Grajek

Thank you for the detailed description. I am sorry that I don't have an immediate answer how to get better surface wind simulations. As I said previously, surface wind simulation is always a challenging issue especially in complex terrain area with high-resolution model run. Very often people turn to post-processing tools to 'correct' the model results. WRF has been frozen and we no longer have resources to develop new schemes and utilities. I am really sorry.
 
Z Grajek said:
View attachment 20670
Hi, I'm using the grid closest to the station. I'm using ASTER 30m data with smooth_passes=3, so the station elevation difference is 1990, and the model is 1908, which seems like a good value to me. The graph shows the mountain station on the left, and the leeward side of the mountains on the right. Is it possible to get something like a pink line on the graph at these high speeds? Because the model clearly doesn't detect when to reduce the speed. I'm also concerned about the strange speed peaks at the station on the leeward side.
Click to expand...
@Z Grajek
Hi Z Grajek,

Could you sorte out this issue please?
 
You must log in or register to reply here.
Back
Top