UCM in WRFV4.6.0: Issue with sf_urban_physics=1

[Shailaja Wasti]

Hi all,

I hope everyone is doing well.
I am running WRFV4.6.0 for the CONUS with a nested domain. The WRF model runs fine with sf_urban_physics=0, sf_urban_physics=2, and sf_urban_physics=3, but for sf_urban_physics=1, the model crashes after providing 4 hours of output for the coarse domain. I checked the .error and .out files, but couldn't find any specific error.

Is there any physics option I need to adjust in order to use the single layer urban canopy model in WRF? I have attached the namelist.input file for reference.

Best,
Shailaja

 

[Ming Chen]

First of all, your namelist.input indicates this is a single-domain run, i.e., max_dom = 1. Can you double check to make sure you sent the correct file?

What is the error message for the failed case with sf_urban_physics = 1 ?

 

[Shailaja Wasti]

Hi Chen,

Thank you for your quick response.

The previously attached namelist.input file is for a single domain. Here, I have attached the namelist files for all three domains (US-Tx-Hou), as well as the .error and .out files.
I checked the .out and .error files and didn’t find any errors that would cause the simulation to crash after 4 hours of output for the coarse domain. `

 

[Ming Chen]

I couldn't find any error message in your rsl files, which is weird because we usually can have some information like CFL violation, segmentation fault, etc. But in your case nothing seems to be output to rsl files.

There are a few namelist files in your namelist.zip, i.e.,

namelist1_WRFCoarse.input1    namelist1c_ndown.input1c    namelist2c_ndown.input2c
namelist1b_Real.input1b        namelist2b_WRF.input2b        namelist3_WRF.input3

Can you explain what they are?

If you run this case using ndown, can you switch to two-way nested run?