cu_physics : on and off for 30km and 10 km domain

[ruchirabiswas]

I am performing a sensitivity study to see the effect of cumulus parameterization (KF Scheme). I am using two nested domains; 30km and 10km (study area). I am performing two experiments where in control, I am running the model with KF Scheme and in test, I am switching it off in both domain. However, when I turn off the cu for both domains, the model is blowing off. I also turned off bl_pbl_phy = 0. Can someone let me know what am I missing? Should I just turn it off for the finer grid since my area of interest is the finer grid? I am attaching the namelist.input with the changes I have made.

 

[kwerner]

Hi,
Yes, you will need cumulus parameterization for coarser domains (typically anything > 10km), so you should turn it on for d01. Domains between ~3km and 10km are considered to be somewhat of a "gray area" for cumulus parameterization - meaning it's not clear whether the clouds will be resolved in that range. You may want to test turning it off for d02, vs turning it on for d02 and see which provides the most realistic results.