[RegCNET] Problems in RegCM nested with CCSM3

[Moetasim]

Hi,
We have published our CCSM-RegCM3 interface results as a part of our recent 
paper. You can check it :
http://www.purdue.edu/eas/earthsystem/Diffenbaugh_GRL_06.pdf
The interface validation is for USA, which includes your domain as well. You 
can see in figure 01 that 'RegCM driven by CCSM + observed SST' has a dry bias 
over texas region, when compared with CRU/CCSM/'RegCM driven by NNRP + observed 
SST'. Your results are consistent with what we have already observed in our 
RegCM-CCSM simulations.

Apart from that, a 50 x 50 domain may also be too small for RCM to give its 
independent solution.

Moet



Quoting Guangyu Wu <wgy1972 at yahoo.com>:

> Hi all,
> 
> I am conducting a 25-year control run (1971-1995)
> using CCSM3 20th Century experiment output data (Q, T,
> U, V, Z3) as the ICBC. The domain is a 50x50, 30km
> spacing grid covering the whole Texas State, U.S..
> 
> I used the codes (CCSM.f)provided with the latest
> RegCM3 to generate the ICBC files. However, the output
> results from RegCM simulations are not good.
> Especially for summer, the monthly domain averaged
> precipitation are often below 0.1 mm/day. I tried Kuo,
> Grell, and MIT scheme and all have the same problem. 
> 
> 
> I checked back the monthly precipitation (PRECC+PRECL)
> outputted from CCSM3 and found this area has around
> monthly averaged 2-3 mm/day precipitation for summer.
> Those are close to the CRU observed values. So the
> problem might be associated with the RegCM3 codes.
> 
> I wonder if someone can give me some tips regarding
> this problem. Any suggestions would be greatly
> appreciated.
> 
> Guangyu Wu
> 
> 
> 
> 
>  
> 
________________________________________________________________________________
____
> Cheap talk?
> Check out Yahoo! Messenger's low PC-to-Phone call rates.
> http://voice.yahoo.com
> _______________________________________________
> RegCNET mailing list
> RegCNET at lists.ictp.it
> https://lists.ictp.it/mailman/listinfo/regcnet
>