I found a constant field for latent heat flux in real case large-eddy simulation but the observation

[batebana]

Dear Sir,
I'm running WRF 3.8 in a real case with 6 domains.
3 domains are mesoscale simulations and inner other 3 domains stand for real case large-eddy simulation (LES).
I have used one-way nested to feed mesoscale simulation from domain 3 to LES d01 (d01 of LES refers to d04)
The namelist.input and some of my results are attached here.
I found a nearly constant value for sensible heat flux (SHF) in LES while in the MSC simulation SHF is comparable to the observation field.

Besides this result, the Latente Heat Flux (LHF) is comparable to the observation field.
All these results are from the same output file (ts file).

The flux-gradient relationship is given as in Monteith and Unsworth (2013), page 152, http://denning.atmos.colostate.edu/readings/Monteith.and.Unsworth.4thEd.pdf,
Skin temperature and 2-meter temperature can be used to compute dT from model output.
We know that Ch varies with time however, we have chosen some fixed value objectively to understand the model output. So when choosing Ch=0.005
we get the SHF result nearly capturing the observation. However, when choosing Ch=0.001 we get the SHF result nearly the constant. We concluded that the model underestimated the Ch which made SHF nearly constant value. ( Please see the attached files)

My first question:
How could I adjust the setting so that the sensible heat can be well simulated in LES?

(Particularly if my comments in the attached document are right, I'm looking for how to change my setting that can enhance the LES simulation result for SHF)

Is there any reference that meets this problem before?

I will highly appreciate your contribution to fixing this issue


M. KPAIKPAI Batebana,
Laboratory of Cloud-Precipitation Physics and severe storm,
Institute of Atmospheric Physics,
University of Chinese Academy of Sciences (CAS),
Beijing - China

 

[kwerner]

This topic has been moved from the "General FYI" section, to the physics section of the forum. Someone will respond to this query soon.

 

[batebana]

This topic has been moved from the "General FYI" section, to the physics section of the forum. Someone will respond to this query soon.

Dear Sir,
I'm still waiting for help, till now there is nothing new from the error I got about sensible heat flux in LES.
I will appreciate any suggestion to help me go forward.

 

[kwerner]

Hi,
I would first like to apologize for the extreme delay in response. It seems that this was overlooked by the person responsible for this subject during the time of your original post. We are very sorry, and thank you for your patience.

1) When you say you are using one-way nesting to run d04, are you referring to using the ndown program? If so, I assume you run d01-d03 together and then use ndown to run from d03 to d04? Correct me if I'm wrong.
2) Have you actually run d05 and d06, as well, or are you stopping at d04 because you already see the problem there?
As an initial test, can you try the current version of WRF (WRFV4.1 or WRFV4.1.1) to see if this is any different? There have been several updates since then, so I'd like to make sure nothing we modified in the code actually corrects this.

Thanks!