Re: taylor diagram

From: Dennis Shea <shea_at_nyahnyahspammersnyahnyah>
Date: Thu Sep 01 2011 - 09:20:57 MDT

Here is an excerpt from a script

[snip]
                 ; temporal variance at each grid point [local]
   vref_var_T = dim_variance_n(rdata, 0 ) ; (lat,lon)
   vcase_var_T = dim_variance_n(cdata, 0 )

                 ; wgted areal *local* temporal variance
   wvar_ref_T = sum(wgt_S*vref_var_T)/sumwgt_S
   wvar_case_T = sum(wgt_S*vcase_var_T)/sumwgt_S

                 ; sqrt of ratio of spatially weighted variances
   wvar_ratio_T = (wvar_case_T/wvar_ref_T)^0.5
[snip]

On 9/1/11 9:02 AM, Sanjiv Kumar wrote:
> Hello,
>
> This may be very basic question. But i think it is better to clarify.
>
> 'In the Taylor Plot (http://www.ncl.ucar.edu/Applications/taylor.shtml)
>
> RATIO is ?
>
> document says: *RATIO:* ratio of the standardized variances
>
> question is - is it ratio of 'STANDARD DEVIATION' or 'VARIANCE'
>
> because 'STANDARD DEVIATION' = (VARIANCE)^0.5, it makes difference
> whether you consider standard deviation or variance.
>
> I think it should be ratio of 'STANDARD DEVIATION' (Taylor, 2001).
>
> Please let me know.
>
> Best Regards
> Sanjiv Kumar
>
>
> _______________________________________________
> 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 Thu Sep 1 09:21:06 2011

This archive was generated by hypermail 2.1.8 : Wed Sep 07 2011 - 10:58:58 MDT