[RegCNET] Problem in running RegCM with 2*CO2

guogl guogl at ieecas.cn
Mon Feb 23 03:45:04 CET 2009


Dear Sir
I am doing a 2*CO2 test with RegCM3, and there are several problem:
For example, I will double CO2 of 2006010100-20070101.
1) There have two ways to double CO2.
    One is changing the number in  BLOCKDATA001.F directly:
I change

data ((cgas(i,j),i=1,6),j=1991,2049) /  ! A2 SCENARIO 
..........................................................
2006.0, 382.97, 1871.4, 323.61, 269.39, 563.25  

 to

data ((cgas(i,j),i=1,6),j=1991,2049) /  ! A2 SCENARIO 
..........................................................
2006.0, 765.94, 1871.4, 323.61, 269.39, 563.25  

The other is changing the line

       co2vmr = cgas(2,lyear)*1.e-6

                to

                co2vmr = cgas(2,lyear)*1.e-6 * 2.

                in radini.F

Are both of methods right? if right, Are their  results of two methods the  same?

2) I have two regcm versions,one is regcm.tar.gz  and the other is regcm_beta_20080711.tar.gz .I find their BLOKDATA001.F are different.
 
A:     Their values are different. for example,
data ((cgas(i,j),i=1,6),j=1991,2049) /  ! A2 SCENARIO 
..........................................................
2006.0, 382.97, 1871.4, 323.61, 269.39, 563.25  
in regcm.tar.gz /../Main/BLOKDATA001.F

and 
data ((cgas(i,j),i=1,6),j=2001,2050) /  ! A2 SCENARIO 
........................................................
2006.00, 378.40,1820.60, 321.40, 269.39, 563.25, 
in  regcm_beta_20080711.tar.gz/../Main/BLOKDATA001.F

why?
B:  BLOKDATA001.F contains 'A1B' '20C' 'A2' 'B1' 'B2' scenarios in regcm_beta_20080711.tar.gz/../Main/.
I don't know what is the meaning of '20C SCENARIO'.

C:  If I want to do a CTRL test of 2006010100-2007010100 with regcm.tar.gz, Is the default value of CO2 'A2 SCENARIO' ?

Expect your help! Thanks!

With Best Regards!
Guo
2009-02-23 



guogl 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.ictp.it/pipermail/regcnet/attachments/20090223/1cef04c4/attachment-0001.html>


More information about the RegCNET mailing list