Re: R: Re: Taylor Diagram

AS noted at http://www.ncl.ucar.edu/Applications/taylor.shtml

"The pertinent statistics are the weighted pattern correlation
(pattern_cor) and the normalized root-mean-square (RMS) differences."

These are both 'spatial quantities'

You are comparing climatologies ... not true time series.

I think the code is giving you waht it can.

Good luck

On 2/19/13 12:01 PM, markus.mingel@libero.it wrote:
> Sorry for being vague...after several attempts the script works, but I am not
> sure it is doing what I want.
>
> My purpose is comparing the mean annual cycle of simulated surface temperature
> from different models with observations.
>
> First, following the script taylor_7.ncl I defined:
>
> SEASONS = (/ "ANN" /)
>
> The script correctly reads the model and data, and both have the dimensions of
> 180x360x12. However after these calls:
>
> cdata = getData( fc, VAR_COMPARE(nv), SEASONS(ns), gw, False )
> tdata = getData( ft, VAR_COMPARE(nv), SEASONS(ns), gw, False )
>
> both cdata and tdata have the dimensions of 180x360, therefore all the
> statistics computed after getData are based on the spatial pattern rather than
> on the comparison of the time series. Or am I wrong and the script is doing
> what I want?
>
> Thanks again for the help
>
> Best
> Markus
>
>
>> ----Messaggio originale----
>> Da: haley@ucar.edu
>> Data: 19/02/2013 17.18
>> A: "markus.mingel@libero.it"<markus.mingel@libero.it>
>> Cc: <ncl-talk@ucar.edu>
>> Ogg: Re: [ncl-talk] Taylor Diagram
>>
>> Markus,
>>
>> As Dennis stated, this question is too vague, Please see our posting
> guidelines for information on how to get help, and how to email us with
> questions about your script:
>>
>> http://www.ncl.ucar.edu/Support/posting_guidelines.shtml
>>
>>
>> --Mary
>>
>> On Feb 19, 2013, at 8:13 AM, markus.mingel@libero.it wrote:
>>
>>> Dear ncl users,
>>>
>>> I am newbie with NCL and maybe it is a trivial question.
>>>
>>> I was wondering how create a Taylor diagram showing the performances of
> some
>>> models in reproducing the mean annual cycle of a reference dataset?!
>>> Specifically, I have a gridded reference temperature of size 180x360x12
> (lat,
>>> lon, time) and 2 models on the same grid: I already have checked the file
>>> taylor_7.ncl but my script is not working.
>>>
>>> Do you have any link or sample script to share?
>>>
>>> Thanks into advance for help
>>>
>>> Markus
>>> _______________________________________________
>>> 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
>
_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Tue Feb 19 22:35:36 2013