[ncl-talk] Options in eofunc

[Yohei Takano]

Dear Dennis,

  Thank you for your reply and I was not clear enough about the example.
I saw the option here in the document (and Example 7 is at the bottom).
  http://www.ncl.ucar.edu/Document/Functions/Built-in/eofunc.shtml

  Somehow my NCL (ver 6.3.0) does not give any warnings and I am not 
sure why... Sounds like it is safe to set optEOF and specify options 
when I conduct EOF analysis.

  Thank you again for clarification.

  Sincerely,

Yohei


On 01/20/2017 04:21 PM, Dennis Shea wrote:
> ???    There is no Example 7
>
> http://www.ncl.ucar.edu/Applications/eof.shtml
>
> re: "I am not 100% sure what the third option (80.) means in this case"
>
> I am not sure either!
>
> ===
> That said, the last argument is prototypes as 'logical
>
> function eofunc (
> 		data    : numeric,
> 		neval   : integer,
> 		optEOF  : logical   <======
> 	)
>
> If the function is invoked as:
>        $ evec= eofunc(ptw,neval,80.)
>
> You should have received the following message
>
> warning:Argument 2 of the current function or procedure was coerced to
> the appropriate type and thus will not change if the function or
> procedure modifies its value
>
> The correct usage is:
>
> optEOF       = True
> optEOF at pcrit = 80
> evec= eofunc(ptw,neval,optEOF)
>
> ==============================
>
> Deep Background: NCL actually coerces (converts) the '80.' to a True.
> However, since there are no optional arguments associated with this
> coerced argument, it would operate as if the 'optEOF-False'
>
>
>
> On Fri, Jan 20, 2017 at 7:52 AM, Yohei Takano
> <yohei.takano at mpimet.mpg.de <mailto:yohei.takano at mpimet.mpg.de>> wrote:
>
>     Dear NCL users,
>
>       I have a question on "eofunc".
>     The NCL website tells me that eofunc has three input (options)
>
>       function eofunc(
>                     data    : numeric,
>                     neval   : integer,
>                     optEOF  : logical
>                     )
>
>       and I understand that you can set options in "optEOF (logical)"
>     such as
>     using correlation matrix and setting the % of non-missing points.
>
>       However, I found a couple of examples using "eofunc" like below,
>
>        $ evec= eofunc(ptw,neval,80.)
>       (Example 7 on the NCL website)
>
>       and I am not 100% sure what the third option (80.) means in this case.
>     I thought the third option supposed to be logical but apparently some
>     of the example contains a number. I thought this is the same as setting
>     the % of non-missing points because when I compared the two methods
>     (i.e. one with setting the optEOF and the other just setting values like
>     in Example 7), I see the same results (EOF) but am I correct on this?
>
>       If anyone can clarify on this that will be great.
>     Thank you very much in advance.
>
>       Sincerely,
>
>     Yohei Takano
>
>     --
>     Yohei Takano Ph.D
>     Max Planck Institute for Meteorology
>     Bundesstraße 53, Room 207
>     20146 Hamburg, Germany
>     Tel: +49 (0)40 41173 - 150 <tel:%2B49%20%280%2940%2041173%20-%20150>
>     Email: yohei.takano at mpimet.mpg.de <mailto:yohei.takano at mpimet.mpg.de>
>     _______________________________________________
>     ncl-talk mailing list
>     ncl-talk at ucar.edu <mailto:ncl-talk at ucar.edu>
>     List instructions, subscriber options, unsubscribe:
>     http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>     <http://mailman.ucar.edu/mailman/listinfo/ncl-talk>
>
>

-- 
Yohei Takano Ph.D
Max Planck Institute for Meteorology
Bundesstraße 53, Room 207
20146 Hamburg, Germany
Tel: +49 (0)40 41173 - 150
Email: yohei.takano at mpimet.mpg.de