Vortex-following: Inner domain does not move

[huyen hoang]

Hi,

I'm trying to use vortex following with 3 domains (18-6-2km resolution), d01 is fixed, d02 and d03 are vortex following.
Set up in my namelist.input:
input_from_hires = .false.,.true.,.true.,
rsmas_data_path = "/media/work/hoanghuyen/wrfv4.2/runWPS/geog/SurfaceFiles"
vortex_interval = 30, 30, 30,
max_vortex_speed = 30, 30, 30,
corral_dist = 8, 15, 20,
track_level = 85000
time_to_move = 10000, 60, 60,
However, d02 and d03 do not move (as in the file attached).

So, I have some questions:
- Domain 1 is fixed, so do I have to set time_to_move as 10000 (higher than running hours) or 0?
- Does the Inner domain work better 12h/ 24h later, or we can set it to run immediately the with outer domain?
- vortex_interval is how often the new vortex position is computed (in minutes), so is it mean that after the vortex position is defined, will have a new domain position?
- corral_dist: is there any formula we can you to define the number of coarse grid cells that the moving nest is allowed to get?

Any help is greatly appreciated.

 

[huyen hoang]

Hi,

I reset up WRF, in the environmental variable test and build libraries, I tried to set TERRAIN_AND_LANDUSE=1 environment variable but was unsucessful.
So I wonder which step is suitable for set TERRAIN_AND_LANDUSE.

Many thanks,

 

[kwerner]

I am going to respond here to your original question. Since the second inquiry is different, will you please post that as a new questions in the forum? It helps to keep topics separate for future searchability purposes. Thanks!

- Domain 1 is fixed, so do I have to set time_to_move as 10000 (higher than running hours) or 0?

I believe you can set this to 0 for domain 01

- Does the Inner domain work better 12h/ 24h later, or we can set it to run immediately the with outer domain?

This can be set to run immediately with the outer domain if you'd like. Otherwise, I'm not sure if 12 or 24 hours later is better. It depends on your application. You could try a short test with both to see which gives you more accurate results.

- vortex_interval is how often the new vortex position is computed (in minutes), so is it mean that after the vortex position is defined, will have a new domain position?

Yes, that is correct.

- corral_dist: is there any formula we can you to define the number of coarse grid cells that the moving nest is allowed to get?

I'm not aware of a specific formula, but a commonly-used/recommended setting is = 8, 15, 15

 

[huyen hoang]

Hi,

I reran other cases
But all of them fail to follow the vortex even when I try WRFv4.4.1, WRFv4.2, or WRFv4.0.
I have attached rsl log, namelist.wps, and namelist.input file here.
Can you take a look and give me some advice,
Thank you,

 

[kwerner]

Hi,
In your rsl.out.0000 file, I see that the nest starts to move at 2020-10-28_13:00:00, and it should only move when the vortex center (or area of lowest pressure) moves so that the edges of that domain would reach the edge of its parent domain (within the number of rows you specified). If it's moving, what is it following, exactly? Can you package up a few of your wrfout* files (as a *.tar) that go through a few time periods past the initial move time and send those to me? If the files are too large to attach, and if you don't have another method of getting those files to me, see the home page of this forum for information on sharing large files. Thanks!

 

[huyen hoang]

Hi,
The uploaded file is too large for the server to process. So I uploaded file wrfout to GG Drive: WRF - Google Drive
Thanks!

 

[kwerner]

Can you package that file as a *.tar instead of *.rar? We aren't able to open .rar files. Thanks, and sorry for the inconvenience.