tcorr function

Mateus S. Teixeira mtex2k3 at YAHOO.COM.BR
Mon Feb 28 06:47:11 EST 2005


Lee, it works very well!
Thanks!!

Mateus.

Em Dom 27 Fev 2005 17:45, you wrote:
> Mateus,
> It looks like a problem with having two files open at the same time that
> have different grid dimensions.  Here is a set of commands that should work
> for your tcorr:
>
> ga> open ic.ctl
> ga> set t 1 37
> ga> define var=ic
> ga> close 1
> ga> open hgt.ctl
> ga> set lon __ __
> ga> set lat __ __  (you fill in)
> ga> set t 1
> ga> d tcorr(var,hgt,t=1,t=37)
>
> In this case you just open the ic file, define the ic for each time, close
> it, and you can use the defined variable (var) in the tcorr expression with
> only the hgt.ctl file opened.
>
> Good luck,
> Lee
>
> On Fri, 25 Feb 2005, Mateus S. Teixeira wrote:
> > Hi,
> > I'm trying to calculate some time correlation fields between two data
> > using TCORR function but something is going wrong.
> > When a give the display command a I have the following error message:
> >
> > TCORR:  dim = 3, start = 1, end = 37
> > Error from TCORR:  1st arg must be 0-D
> > Error from TCORR:  Error getting grids
> > Operation Error:  Error from tcorr function
> >   Error ocurred at column 1
> > DISPLAY error:  Invalid expression
> >   Expression = tcorr(ic.2,hgt.1,t=1,t=37)
> >
> > IC.2 is a time series of single data per time and HGT.1 is a time series
> > of grided data. It seems that TCORR function try to read the data from
> > ic.2 variable but the dimensions are not correct.
> >
> > My CTL files for these data is:
> >
> > DSET ^ic.dat
> > TITLE Precipitation Index
> > UNDEF -9.99E33
> > XDEF 1 LINEAR 180 1
> > YDEF 1 LINEAR -65 1
> > ZDEF 1 LINEAR 1 1
> > TDEF 37 LINEAR 12Z01OCT2001 1YR
> > VARS 1
> > ic 0 99 prec. index
> > ENDVARS
> >
> > DSET ^djf.D0.%y2.hgt.500mb.dat
> > OPTIONS template
> > TITLE Geop. Height data
> > UNDEF -9.99E33
> > XDEF 72 LINEAR 180.00 2.5
> > YDEF 27 LINEAR -65.00 2.5
> > ZDEF 1 LINEAR 1 1
> > TDEF 37 LINEAR 12Z01OCT2001 1YR
> > VARS 1
> > hgt 0 99 geopotential height
> > ENDVARS
> >
> > Thanks.
> > Mateus.



More information about the gradsusr mailing list