[gradsusr] any help/thoughts?

[krishnamohan]

Hello Dr. Muthuvel Chelliah,

I assume that you require a new variable such that, in a particular year,
you want values for that variable in a some months ( Say June to Septmber)
and have to set all other months to undefined. I think you can do this in a
tricky way. Please try the attached script. I will explain the concept here
with HadISST data. This method can only be used for monthly data since 'modify'
command does not support creating daily climatology.

'reinit'
'sdfopen HadISST_sst.nc'

suppose you want only data for the indian summer monsoon time ( June to
September). ie months 6 to 9

'set t 6 9'
'define sstclim = ave(sst, t+0, t=24, 12)'
'modify sstclim seasonal'

now this new variable sstclim contains only values for June to September
and all other values undefined. Now create a new variable such that the
values are a constant field '1' form June to September months and all other
months are undefined.

'define clim=sstclim/sstclim'
'modify clim seasonal'

This new variable contains value 1 from June to September and all other
values undefined. Now multiply it with the real sst data for the entire
time period.

'set t 1 100'
'define sstnew=sst*clim'

The new variable sstnew will have values only for June to September in
every year.

Krishnamohan


On Tue, Apr 16, 2013 at 6:30 AM, Jennifer Adams <jma at cola.iges.org> wrote:

> Take a look at the docs for the tmave() function. You may not want
> averaging over your time series, but it might be helpful in setting up a
> mask to apply to your longer time series to get rid of periods you are not
> interested in. You could create an ascii file with 0's and 1's, convert to
> netcdf with ncgen, and apply (multiply or whatever) that mask to your whole
> time series.
> --Jennifer
>
>
> On Apr 15, 2013, at 5:14 PM, Muthuvel Chelliah - NOAA Federal wrote:
>
> Hello Grads experts,
>
> Is there any recommendation for the following situation.
>
> After I define a variable VAR (assume daily data) for a period of
> long time, can I selectively choose a shorter sub-period within this full
> period and set VAR to undef value, and still
> use the full period of VAR values, with this newly set undefined
> value for that sub period?  I hope this is not confusing.
>
> The problem I run into is, when I try to redefine VAR
> for the shorter period, full period VAR gets deleted, and a new VAR
> is created only for the shorter period.
>
> Thanks for any help/suggestions.
>
> Muthu
>
> ------------------------------------------------------------------
>
> Dr.Muthuvel Chelliah, eMail:Muthuvel.Chelliah at noaa.gov
> Rm.3008, Climate Prediction Center, Ph:(301) 683-3393
> NCEP/NWS/NOAA/U.S.Dept.of Commerce,
>
> 5830, University Research Court, NCWCP
> College Park, MD 20740.
>
> _______________________________________________
>
> gradsusr mailing list
> gradsusr at gradsusr.org
> http://gradsusr.org/mailman/listinfo/gradsusr
>
>
> --
> Jennifer M. Adams
> IGES/COLA
> 4041 Powder Mill Road, Suite 302
> Calverton, MD 20705
> jma at cola.iges.org
>
>
>
>
> _______________________________________________
> gradsusr mailing list
> gradsusr at gradsusr.org
> http://gradsusr.org/mailman/listinfo/gradsusr
>
>


-- 
Krishnamohan.K.S
CSIR Research Fellow
Department Of Atmospheric Sciences
Cochin University of Science and Technology
Cochin,India
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://gradsusr.org/pipermail/gradsusr/attachments/20130416/99c3ddc3/attachment-0003.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: krish.gs
Type: application/octet-stream
Size: 278 bytes
Desc: not available
Url : http://gradsusr.org/pipermail/gradsusr/attachments/20130416/99c3ddc3/attachment-0003.obj