[R-sig-Geo] Constraining movements in adehabitat

[Luca Borger]

Hi Tim,

as suggested by David the methods developed by Jason Matthiopolous would be the most adequate for your problem, but no one has implemented them in R (more on this below). An alternative useful one could be the so-called non-parametric kernel method of Getz et al.:

Getz WM, Fortmann-Roe S, Cross PC, Lyons AJ, Ryan SJ, et al. 2007 LoCoH: Nonparameteric Kernel Methods for Constructing Home Ranges and Utilization Distributions. PLoS ONE 2(2): e207. doi:10.1371/journal.pone.0000207 

also called local convex-hull home range method (LoCoH):

Getz, W. M. and C. C. Wilmers, 2004. A local nearest-neighbor convex-hull construction of home ranges and utilization distributions. Ecography 27:489-505

The method is especially well-suited to follow sharp habitat boundaries and exclude non-used or unavailable areas. Of course, how well it will work in your case will depend on how well you have sampled the movements of your study species, i.e. if your sampling regime (intensity and timing of sampling) is adequate for the movement characteristics of your population and the complexity of the landscape. The method is implemented in adehabitat/ArcGis, but a nice extension to run it in ArcView 3.x is also available from the first author's website. 


Given that we are on the r-sig-geo list, the two papers by Jason Matthiopolous mentioned by David are among the best contributions so far written regarding animal movements and home ranges, so if anyone would be willing and able (and have the time) to implement these methods in R I believe it would be very useful to many people ;-)



HTH


Cheers,


Luca


--------------------------------------------
Luca Borger, PhD
Postdoctoral Research Fellow
University of Guelph
Guelph, ON, Canada

email: lborger at uoguelph.ca
http://www.researcherid.com/rid/C-6003-2008
http://uoguelph.academia.edu/LucaBorger
--------------------------------------------


----- Messaggio originale -----
Da: r-sig-geo-request at stat.math.ethz.ch
A: r-sig-geo at stat.math.ethz.ch
Inviato: Giovedì, 14 maggio 2009 6:00:05 GMT -05:00 U.S.A./Canada, stati orientali
Oggetto: R-sig-Geo Digest, Vol 69, Issue 14

Send R-sig-Geo mailing list submissions to
	r-sig-geo at stat.math.ethz.ch

Message: 16
Date: Thu, 14 May 2009 09:18:30 +0200
From: Pinaud David <pinaud at cebc.cnrs.fr>
Subject: Re: [R-sig-Geo] Constraining movements in adehabitat
To: Tim Clark <mudiver1200 at yahoo.com>, R-sig-Geo at stat.math.ethz.ch
Message-ID: <4A0BC5C6.9000005 at cebc.cnrs.fr>
Content-Type: text/plain; charset="windows-1252"; Format="flowed"

Dear Tim,

You can try several papers from Jason Matthioploulos, among:
Matthiopoulos (2003) The use of space by animals as a function of 
accessibility and
preference - Ecological Modelling 159 239-268
Matthiopoulos, J. 2003. Model-supervised kernel smoothing for the 
estimation of
spatial usage. ? Oikos 102: 367?377.
Unfortunately, I don't know a R implementation to do this directly. One 
(quick and dirty) way could be to weight the output of the Kernel 
estimation by a raster representing distance to coast (or ocean/continent)?

Hope it helps

David

Tim Clark a ?crit :
> Dear List,
>
> Sorry, this is a second attempt to post the same message.  My last one got scrubbed so I changed browser options.  Hopefully this one will make it!
>
> I am using the package adehabitat to analyze the movement patterns and home range of the manta ray (Manta birostris) in Hawaii.  The manta rays in this study stayed close to shore, often feeding along the shore line or in small bays.  Since they can not go on land (being fish) their home range is obviously limited to the ocean area.  In calculating home range statistics, is there a way to limit the avalible area to the ocean.  I have read through the papers on Adehabitat but have not found a way to limit the area. The closest I have found is the function buffer().  Have I missed something?  Is there a way to co

> nstrain the movements so that kernels will only be computed for data points located in the ocean?
>
> I appreciate your help and guidence.
>
> Tim
>
>
>
> Tim Clark
> PhD Student 
> University of Hawaii
> Honolulu, HI 96816
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>
>
> __________ Information from ESET Mail Security, version of virus signature database 4073 (20090513) __________
>
> The message was checked by ESET Mail Security.
> http://www.eset.com
>
>
>
>
>
>   

-- 
***************************************************
David PINAUD
Ing?nieur de Recherche "Analyses spatiales"

Centre d'Etudes Biologiques de Chiz? - CNRS UPR1934
79360 Villiers-en-Bois, France 
poste 485
Tel: +33 (0)5.49.09.35.58
Fax: +33 (0)5.49.09.65.26
http://www.cebc.cnrs.fr/