Re: Taylor Diagram

From: Dennis Shea <shea_at_nyahnyahspammersnyahnyah>
Date: Tue Nov 15 2011 - 09:40:57 MST

The 'problem' with daily data is that it can be very noisy.
I am not aware of any use of Taylor diagrams with daily data.

The Taylor diagram requires [1] a pattern correlation.
[2] computation of variances ... this requires (x(i)-xbar)^2
     What to use for xbar?

On 11/15/11 9:31 AM, I N wrote:
> Dear Dennis,
>
> Thanks for the quick reply. My first question was not actually
> about reference grid rather it was about temporal resolution of the
> data.
> It is recommended to use monthly/seasonal data or one can also use
> daily data for taylor diagram?
>
> Regards
> Nadeem
>
>
> On Tue, Nov 15, 2011 at 5:01 PM, Dennis Shea<shea@ucar.edu> wrote:
>> [1] You can use any reference grid. As to what is appropriate for
>> one year of daily data, I don't know. You have to make that
>> decision.
>>
>> [2] for only an 8 degree latitude span, the results will be
>> more or less the same. Again, you shuld make that choice.
>>
>> Good luck
>>
>> On 11/15/11 8:37 AM, I N wrote:
>>>
>>> Dear All,
>>> I have a one year daily precipitation data from different setups of
>>> the same RCMl and observed precipication, all on the same curvilinear
>>> grid (Lambert conformal projection).
>>> The task is to create taylor diagram for whole year, Winter, Summer
>>> and 4 seasons (DJF,MAM,....).
>>> The latitude span of domain is from 42N to 50N (8 degree)
>>>
>>> I have few questions:
>>>
>>> 1) From NCL website, second line on the page "Special Topics: Taylor
>>> Diagrams"
>>>
>>> "The plotted values are generally derived from monthly or seasonal
>>> climatological means of one or more variables"
>>>
>>> This means that it is recommended to use monthly/seasonal climatology.
>>> Does it apply to all cases or in my case (only 1 year of data)
>>> I can use daily data. What is advantage/disadvantage of using daily data?
>>>
>>>
>>> 2) From ncl-talk archive of May 21, 2010.
>>>
>>>>> If the region being tested is 'large' then a weight variable [wgt]
>>>>> should be created. If your region is 'small', you can ignore the
>>>>> weight [wgt=1]. Given that you are looking at a very small area, you
>>>>> can ignore any weighting. Also you have masked out all points outside
>>>>> your area of interest.
>>>
>>> Small and large are relative terms!! If the latitude span is 8 degree,
>>> it is considered small or not?
>>> I find significant differences between wgt=1 and
>>> wgt=cos(lat2d*0.01745329) versions of taylor diagram.
>>>
>>> Thanks in advance
>>> Regards
>>> Nadeem
>>> _______________________________________________
>>> 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 Nov 15 09:41:11 2011

This archive was generated by hypermail 2.1.8 : Thu Nov 17 2011 - 11:50:12 MST