[AniMov] kerneloverlap() UDOI in adehabitatHR (Morellet Nicolas) (marc fernandez)

marc fernandez marc.fern at gmail.com
Thu Aug 1 19:48:31 CEST 2013


Dear all,

I have been looking at my results and there is something wrong in the
results of the UDOI indexes. Using the BA index things seem to work
properly, ranging form 0.7 to 1 (I have strong overlaps).

But the UDOI is strange. If I understand it properly the overlap of an
animal with itself should be 1. But in this case I get values around 2.5
which according to Fierberg (2005) could mean that they have high degree of
overlap but they are nouniformly distributed...which in this case doesn't
make any sense, as is the same animal.

Can anyone help me?

Thanks

Marc

P.S:I attach here the new sample file without outliers


2013/7/31 <animov-request at faunalia.it>

> Send AniMov mailing list submissions to
>         animov at faunalia.it
>
> To subscribe or unsubscribe via the World Wide Web, visit
>         http://lists.faunalia.it/cgi-bin/mailman/listinfo/animov
> or, via email, send a message with subject or body 'help' to
>         animov-request at faunalia.it
>
> You can reach the person managing the list at
>         animov-owner at faunalia.it
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of AniMov digest..."
>
>
> Today's Topics:
>
>    1. Re: kerneloverlap() UDOI in adehabitatHR (Morellet Nicolas)
>       (marc fernandez)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Tue, 30 Jul 2013 21:14:33 +0000
> From: marc fernandez <marc.fern at gmail.com>
> To: animov at faunalia.it
> Subject: Re: [AniMov] kerneloverlap() UDOI in adehabitatHR (Morellet
>         Nicolas)
> Message-ID:
>         <CAOVqF++Y=ipREcRNvD==
> jR2GT0rVHGjRYAbRyt_ocbJzX+FhRQ at mail.gmail.com>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Dear Nicolas,
>
> Thanks for your answer. You are totally right.
>
> I am in the middle of the sampling campaign and I spend a lot of hours at
> sea, then when I sit to do the analysis sometimes I make this silly
> mistakes. In this case I chosed the wrong files (raw data), with real
> outliers. I selected the right files and now things seems to work better.
>
> I still have to see if the results makes sense in an ecological way.
>
> Thanks a lot for your suggestion.
>
> Cheers,
> Marc
>
>
> 2013/7/24 <animov-request at faunalia.it>
>
> > Send AniMov mailing list submissions to
> >         animov at faunalia.it
> >
> > To subscribe or unsubscribe via the World Wide Web, visit
> >         http://lists.faunalia.it/cgi-bin/mailman/listinfo/animov
> > or, via email, send a message with subject or body 'help' to
> >         animov-request at faunalia.it
> >
> > You can reach the person managing the list at
> >         animov-owner at faunalia.it
> >
> > When replying, please edit your Subject line so it is more specific
> > than "Re: Contents of AniMov digest..."
> >
> >
> > Today's Topics:
> >
> >    1. Re: kerneloverlap() UDOI in adehabitatHR (Morellet Nicolas)
> >
> >
> > ----------------------------------------------------------------------
> >
> > Message: 1
> > Date: Wed, 24 Jul 2013 11:35:02 +0200
> > From: Morellet Nicolas <Nicolas.Morellet at toulouse.inra.fr>
> > To: Animal Movement <animov at faunalia.it>
> > Subject: Re: [AniMov] kerneloverlap() UDOI in adehabitatHR
> > Message-ID: <51EF9FC6.1080608 at toulouse.inra.fr>
> > Content-Type: text/plain; charset="iso-8859-1"; Format="flowed"
> >
> > Hi Mark,
> >
> > I think you should check a bit your data. I have the feeling that there
> > are several "outliers". You can see it easily with:
> >   plot(tot, col=tot$Ind)
> >
> >   Maybe the presence of these outliers induce these strange results,
> > even if I do not know exactly what is behind this UDOI measure.
> >
> > Try for example:
> > overlap<-kerneloverlap(tot[,2], meth="UDOI",percent=50)
> >
> > Results are really different (for the core area).
> >
> > Cheers,
> > Nicolas.
> >
> > Le 23/07/2013 16:29, marc fernandez a ?crit :
> > > Hi,
> > >
> > > I am having some issues in applying the home range overlap metrics
> > > with the adehabitatHR package. I tried to use the UDOI method and the
> > > BA. While the BA seems to work quite well,the values from the UDOI are
> > > very strange UDOI, ranging from 300 to 0.3. I wonder if it could be
> > > solved including the grid size? Or maybe it is related with the
> > > selection of the smoothing parameter? Can anyone help me?
> > >
> > > Here you have the code, and I also attach a  reduced sample of the
> data.
> > >
> > > > tot<-read.csv("totall.csv", header=TRUE, sep=";")
> > > > coordinates(tot) <- ~Easting+Northing
> > > > summary(tot)
> > > Object of class SpatialPointsDataFrame
> > > Coordinates:
> > >              min       max
> > > Easting   268511  576049.4
> > > Northing 4209927 4264506.2
> > > Is projected: NA
> > > proj4string : [NA]
> > > Number of points: 1210
> > > Data attributes:
> > >        date           Ind      Sex
> > >  8/31/05 :  16   s5a    : 77   F:647
> > >  10/2/07 :  15   s9a    : 75   M:563
> > >  7/27/04 :  15   s6b    : 74
> > >  7/11/07 :  13   s8j    : 63
> > >  8/9/05  :  13   s7b    : 62
> > >  10/13/07:  12   m2c    : 59
> > >  (Other) :1126   (Other):800
> > > > overlap<-kerneloverlap(tot[,2], meth="UDOI")
> > >
> > > Thanks a lot!
> > >
> > > Cheers,
> > >
> > > MArc
> > >
> > >
> > > _______________________________________________
> > > AniMov mailing list
> > > AniMov at faunalia.it
> > > http://lists.faunalia.it/cgi-bin/mailman/listinfo/animov
> >
> >
> > --
> >
> >
> > **********************************************************
> >                     Nicolas MORELLET
> > I.N.R.A. - Centre de Recherches de Toulouse
> > Comportement et ?cologie de la Faune Sauvage (C.E.F.S.)
> > BP 52627 - 31326 Castanet-Tolosan cedex (France)
> >                Tel :  05.61.28.51.28
> >                Fax :  05.61.28.55.00
> > **********************************************************
> >
> > -------------- next part --------------
> > An HTML attachment was scrubbed...
> > URL: <
> >
> http://lists.faunalia.it/pipermail/animov/attachments/20130724/586fd35b/attachment-0001.html
> > >
> >
> > ------------------------------
> >
> > Subject: Digest Footer
> >
> > _______________________________________________
> > AniMov mailing list
> > AniMov at faunalia.it
> > http://lists.faunalia.it/cgi-bin/mailman/listinfo/animov
> >
> >
> > ------------------------------
> >
> > End of AniMov Digest, Vol 89, Issue 3
> > *************************************
> >
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <
> http://lists.faunalia.it/pipermail/animov/attachments/20130730/1b6e422e/attachment-0001.html
> >
>
> ------------------------------
>
> Subject: Digest Footer
>
> _______________________________________________
> AniMov mailing list
> AniMov at faunalia.it
> http://lists.faunalia.it/cgi-bin/mailman/listinfo/animov
>
>
> ------------------------------
>
> End of AniMov Digest, Vol 89, Issue 9
> *************************************
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.faunalia.it/pipermail/animov/attachments/20130801/bd73f932/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: teste.csv
Type: text/csv
Size: 33454 bytes
Desc: not available
URL: <http://lists.faunalia.it/pipermail/animov/attachments/20130801/bd73f932/attachment-0001.csv>


More information about the AniMov mailing list