[ncl-talk] Complex EOF in NCL

[Dennis Shea]

'eofunc' uses a covariance matrix as the basis for deriving EOFs. Consider
two grid-points (stations) denoted by X and Y which have NT time steps.

   cov = SUM [(X(t)-Xave)*(Y(t)-Yave)]/NT

If either X(t) or Y(t) is missing it is not used in the product and NT
is reduced appropriately.

If NT=100 and there are (say) 5 cases where X(t) or Y(t) is missing
then NT= 95 (=100-5)

>From the documentation:

optEOF at pcrit = %: a float value that indicates the percentage of
non-missing points that must exist at any single point in order to be
included in the calculations. The default is 50%. Points that contain more
than 'pcrit' missing values
<http://www.ncl.ucar.edu/Document/Language/fillval.shtml> will be excluded
from the computations.

HTH

On Thu, Feb 2, 2017 at 9:52 PM, Shikha SIngh <08singhshikha at gmail.com>
wrote:

> Hi,
>
> It is possible to do it in MATLAB, but the missing values(NaN) create a
> problem. I am searching for a hack around it.
> Any idea how NCL 'eofunc_Wrap' handles missing values?
>
> Thanks,
>
> On Fri, Feb 3, 2017 at 9:58 AM, saurabh rathore <rohitsrb2020 at gmail.com>
> wrote:
>
>> Hi Shikha
>>
>> There is straight forward MATLAB function for this analysis please give
>> it a try I myself don't know MATLAB but I am confident that what you want
>> is in MATLAB. Please update us also if you are able to solve your problem.
>>
>> On Feb 3, 2017 3:26 PM, "Dennis Shea" <shea at ucar.edu> wrote:
>>
>>> Sorry, no.
>>>
>>> There are several Matlab , search the WWW.
>>> There are also fortran codes
>>>
>>> Good luck
>>>
>>>
>>>
>>> On Thu, Feb 2, 2017 at 9:13 PM, Shikha SIngh <08singhshikha at gmail.com>
>>> wrote:
>>>
>>>> Dear all,
>>>>
>>>> Does NCL have a function to do Hilbert transform and complex EOF of a
>>>> dataset?
>>>>
>>>>
>>>> --
>>>> Regards,
>>>>
>>>> Shikha Singh
>>>> Scientist B,
>>>> Indian Institute of Tropical Meteorology,
>>>> Pune.
>>>>
>>>> +91 9422272531 <+91%2094222%2072531>
>>>>
>>>> _______________________________________________
>>>> ncl-talk mailing list
>>>> ncl-talk at ucar.edu
>>>> List instructions, subscriber options, unsubscribe:
>>>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>>>>
>>>>
>>>
>>> _______________________________________________
>>> ncl-talk mailing list
>>> ncl-talk at ucar.edu
>>> List instructions, subscriber options, unsubscribe:
>>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>>>
>>>
>
>
> --
> Regards,
>
> Shikha Singh
> Scientist B,
> Indian Institute of Tropical Meteorology,
> Pune, India
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20170203/eb804a05/attachment.html