[Thread Prev][Thread Next][Index]

Re: Composite image based on irregular months



> Yun.Li@csiro.au wrote:
>
> > I am trying to make a composite image based on some irregular months
> > of an index. For example, I want a composite image of sea level
> > pressure (SLP) based on irregular months such as L=409, 505, 576 and
> > 625 from the following data in ferret.
> >
> > 2> /data/necp/slp_mon_1948_2002
> > name     title       I         J         K         L
> >  SLP      slp        1:144     1:73      1:1       1:660
> >
> > So far, I can only use the following awkward way to make it, that is,
> >
> > Let P=(slp[d=2, L=409] + slp[d=2, L=505] + slp[d=2, L=576] + slp[d=2, L=625])/4
> > Fill P

Hi Yun & Ansley,

The @WEQ ("weighted equal") transformation is great for building such
composites.  To get the SST of months {2,4,8,11} from COADS:

use coads_climatology
let keys = zsequence({2,4,8,11})
let zero_at_keys = l[gl=sst] - keys
let integrand = zero_at_keys[l=@weq:0]
let sst_masked = sst * integrand[k=@sum]

The composite is then the time-average of sst_masked.  Here's a plot of
the various stages:

set window 1; set mode interp; can v
set v ul; shade/title="INTEGRAND" integrand
set v ur; plot/step=con/title="L_MASK" integrand[k=@sum]
set v ll; shade/y=0/title="SST_MASKED" sst_masked
set v lr; shade/title="SST_COMPOSITE" sst_masked[t=@ave]

Cheers,

Andrew

+--------------------------------------------------------+
|   Dr. Andrew T. Wittenberg   |        GFDL/NOAA        |
|  Andrew.Wittenberg@noaa.gov  |      Princeton, NJ      |
+--------------------------------------------------------+



[Thread Prev][Thread Next][Index]

Dept of Commerce / NOAA / OAR / PMEL / TMAP

Contact Us | Privacy Policy | Disclaimer | Accessibility Statement