Re: covcorm function output

Hi Will,

This indeed looks like a bug. I'm looking into it to see if there's a work-around.

I've filed a ticket, NCL-1681.

--Mary

On Mar 13, 2013, at 8:37 PM, Will Hobbs wrote:

> Hi
>
> I am trying to calculate a 2D covariance matrix using the covcorm() function for an array 'x' dimensioned (time=1872, lon=360). By setting the second element of the 'iopt' argument (see below) I am expecting to get a matrix dimensioned (/nlon,nlon/) , but instead I get a 1D matrix of size (nlon*(nlon+1)/2), as if iopt(1) were set to 0.
>
>
>
> > iopt = (/0,1/)
> > C = covcorm(x(lon|:,time|:), iopt)
> >printVarSummary(C)
>
> Variable: C
> Type: float
> Total Size: 259920 bytes
> 64980 values
> Number of Dimensions: 1
> Dimensions and sizes: [64980]
> Coordinates:
> Number Of Attributes: 2
> _FillValue : 1e+20
> trace : 360
>
>
> Am I missing something?
>
> Many thanks
>
> Will
> _______________________________________________
> ncl-talk mailing list
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk

_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Mon Mar 18 10:35:01 2013