[RegCNET] The other option of PostProc
XUNQIANG BI
bixq at ictp.it
Fri Jun 15 10:41:56 CEST 2007
Dear All:
There is almost no reply from myself about PostProc issue, the
reason is that I have no experience of using the original PostProc
and I use the other codes to do the daily mean and monthly mean.
I have put These codes under the directory of
RegCM/Commons/DAY_MONTH in the latest web version.
If you don't want to convert model output to netCDF format and
you just use GrADS as the graphic tool, AND YOU HAVE PROBLEM OF
USING THE ORIGINAL POSTPROC, you are welcome to use these
codes, first do the daily mean, then do the monthly mean.
The way of using these codes is quite the same as using
SIGMAtoP.f,
Compared with the original PostProc, these codes are not that
user-friendly, you'd better read and understand these codes
(they are very simple codes), then you need:
1. you need edit the code and set im and jn (cut 2 at both sides)
2. pay attention to the following line in daily mean file
(day_ATM.f, day_CHE.f, day_RAD.f, day_SRF.f and day_ICBC.f)
if(month.eq.1) nrec=nvar
if you initial the run from January 1st, use the above lines,
otherwise you need change above line in proper way.
3. pay attention to lines
do mday=1,4
and
c(i,j,n)=c(i,j,n)/4.
(in day_ATM.f, day_CHE.f, day_RAD.f, and day_ICBC.f)
here 4 need to be changed if your save frequency is not every
6 hour.
and in day_SRF.f,
do mday=1,24
and
c(i,j,n)=c(i,j,n)/24.
here 24 need to be changed if your save frequency is not every
1 hour (let's say you save SRF fields every 3 hour, then you need
change 24 to 8).
I plan to let these codes write out the GrADS descriptor file
later to make it more user-friendly.
If you are user of FERRET and NCL, please keep using the
original PostProc.
Best regards,
Bi Xunqiang
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Dr. Xunqiang Bi email:bixq at ictp.it
Earth System Physics Group
The Abdus Salam ICTP
Strada Costiera, 11
P.O. BOX 586, 34100 Trieste, ITALY
Tel: +39-040-2240302 Fax: +39-040-2240449
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
More information about the RegCNET
mailing list