# [AniMov] Problem with kerneloverlap output when method=HR

Mathieu Basille basille.web at ase-research.org
Tue Aug 12 15:16:52 CEST 2014

```Dear Elodie,

Who said that overlaps had to be symmetrical? Based on your data, this
should make it clearer:

coordinates(locs) <- ~Longitude+Latitude

mcp1 <- mcp(locs[, 1])
plot(mcp1)

As you can see, the MCP of the first trip is (almost) entirely included in
the MCP of the second trip. But the reverse is not true.

Also please check the ref from the doc [1], where you will find why, and
which is the best overlap estimate that suits your needs (hint: MCP is
generally not).

Last but not least, not sure that lat/long degrees makes sense when it
comes to overlap... You might want to project your data in some local
coordinate system.

Best,
Mathieu.

[1] Fieberg, J. and Kochanny, C.O. (2005) Quantifying home-range overlap:
the importance of the utilization distribution. Journal of Wildlife
Management, 69:1346-1359.

Le 11/08/2014 02:40, Elodie Camprasse a écrit :
> G’day folks,
>
> I am encountering a problem when working with the function kerneloverlap in
> the package adehabitatHR. I am trying to calculate the overlap in home
> ranges within consecutive trips of the same animal. In order to use the
> function, I first created the spatialPointsDataFrame object called example
> from a dataframe that has in order, a column with the ID (trip number), one
> column with Longitude, one column with Latitude and a column with a
> timestamp (see attached csv) like this:
>
> example_df<-SpatialPointsDataFrame(coords=example[, c(2,3)], data=example[,
> c(1,4)], proj4string = CRS("+proj=longlat +ellps=WGS84")).
>
>  From the output I get (below), I deduce that creating this new object has
> worked.
>
> class       : SpatialPointsDataFrame
>
> features    : 130
>
> coord. ref. : +proj=longlat +ellps=WGS84
>
> variables   : 2
>
> names       :       id,        timestamp
>
> min values  : 1st_trip, 15/10/2013 18:18
>
> max values  : 2nd_trip,  18/10/2013 6:54
>
> When I try to use the kerneloverlap function however, I get an output that
> I don’t understand.
>
>
>
> kerneloverlap(example_df[,1], meth="HR")
>
>            1st_trip 2nd_trip
>
> 1st_trip 1.0000000        1
>
> 2nd_trip 0.4895288        1
>
> I am not sure why the results are not symmetrical (in one instance the overlap between trip 1 and 2 is 1, in the other it is 0.4895288…). Trying the first calculate the UD distribution with kernelUD and using the function kerneloverlaphr to calculate the overlap yielded the same result. I thought that would have fixed the problem as I read on the forums that inconsistent output could come from the grid size changing from individual to individual (or in my case trip to trip) but using the parameter same4all=TRUE and conditional=TRUE for kerneloverlaphr did not provide a different result.
>
> Could someone let me know if I have made a mistake and how to fix it
>
> Kind regards,
>
> Elodie
>
> Elodie Camprasse
>
>
> Hawthorn, VIC 3122
>
> Australia
>
> Email: elodie.camprasse at gmail.com <mailto:elodie.camprasse at gmail.com>
>
> Website: http://hors-des-sentiers-battus.e-monsite.com/
>
> Mobile: (+61) 049 794 0793 <tel:%28%2B61%29%20049%20794%200793>
>
>
>
> _______________________________________________
> AniMov mailing list
> AniMov at faunalia.it
> http://lists.faunalia.it/cgi-bin/mailman/listinfo/animov
>

--

~\$ whoami
Mathieu Basille, PhD

~\$ locate --details
University of Florida \\
Fort Lauderdale Research and Education Center
(+1) 954-577-6314
http://ase-research.org/basille

~\$ fortune
« Le tout est de tout dire, et je manque de mots
Et je manque de temps, et je manque d'audace. »
-- Paul Éluard

```