Re: Outlier from Satelite Data

From: Dennis Shea <shea_at_nyahnyahspammersnyahnyah>
Date: Wed Dec 12 2012 - 16:24:38 MST

Further, you said you use stat_dispersion .... that provides
you with assorted measures of the variable's distribution.
This includes robust measures, eg *tile

On 12/12/12 4:21 PM, Will Hobbs wrote:
> Xi
>
> You've answered your own question; you can use the 'where()' function to remove any values that are outside an acceptable range.
>
> How you decide what that acceptable range is will depend on your knowledge about the particular region, datasets and so on, is not really a question about NCL at all.
>
> (That said, I would respectfully suggest that a negative value of precip is wrong, so you could always set zero to your lower bound…..)
>
> Will
>
> From: Xi Chang <xi.chang01@gmail.com<mailto:xi.chang01@gmail.com>>
> Date: Thursday, 13 December 2012 10:09 AM
> To: "ncl-talk@ucar.edu<mailto:ncl-talk@ucar.edu>" <ncl-talk@ucar.edu<mailto:ncl-talk@ucar.edu>>
> Subject: Outlier from Satelite Data
>
> Hallo NCL,
> I am using precipitation dataset derived from Satellite estimate. When i check my data using stat_dispers i found that the max precip within my data over 300 mm/day. Is there any method how to remove the outlier ( how to determine them) ? i found in the last discussion, by setting something like this in the command, how to select the suitable range in this case.
>
> xBIG = ...
> xSMALL = ...
> x = where (x.lt.xSMALL .or x.gt.xBIG, x@_FillValue, x)
>
> Thanks a lot.
>
> Xi
>
>
>
>
> _______________________________________________
> 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 Dec 13 08:25:00 2012

This archive was generated by hypermail 2.1.8 : Wed Dec 12 2012 - 19:07:01 MST