Wind flow at the inlet (Complex terrain)

[Ella Chaumont]

Hello all,

I am running some simulations and trying to get a developed wind field over a complex terrain. I use 'ATMOSPHERIC' profile along with N_EDDY, L_EDDY, VEL_RMS. 

Ex:

&SURF ID='INLET', VEL=-5.0, PROFILE='ATMOSPHERIC', Z0=10.0, PLE=0.143/

&VENT PBX= 0.0, SURF_ID='INLET', N_EDDY=60, L_EDDY=6, VEL_RMS= 0.5/

I have a complex terrain and it starts at the inlet. Please see the bellow image. As shown in the image, I can see the velocity at the inlet is abnormal (I see it as like hitting the terrain which is made with solid obstacles) and the velocity is very high (around 130 m/s max as shown in the Velocity colorbar). 

 Could you please give me an idea to avoid this effect? Thank you in advance.

[dr_jfloyd]

In the image, it looks like you have a small gap between the terrain and the -x boundary. The SURF imposes a fixed velocity profile at the boundary. With gas cells at the boundary, air will enter the domain there and it only has once place to go -- up until it encouters to region above the terrain where the upward jet bends over due to inflow.

[David Sheppard]

Once you have a model with an established velocity field over the entire domain, you should consider saving the velocity field using DT_UVW.  Then you can initialize subsequent runs with the same velocity field using UVWFILE. 

If you only define the conditions at the boundaries, you have to wait for the velocity field to propagate throughout the domain before you start your scenarios.