Species co-occurrence models

158 views
Skip to first unread message

Jorge A. Ahumada

unread,
Sep 19, 2011, 3:36:39 PM9/19/11
to Unmarked package
Anybody knows if unmarked can handle species co-occurrence models? i.e. Model the occurrence of one species as a function of the occurrence of another species..

Thanks,

Jorge


--
Jorge A. Ahumada, PhD
Technical Director, TEAM Network
Science and Knowledge Division
Conservation International
2011 Crystal Dr. Suite 500
Arlington, VA 22202
Office: +1-703-341-2543
Mobile: +1-202-716-3374
http://www.teamnetwork.org/


 

Kery Marc

unread,
Sep 19, 2011, 3:49:31 PM9/19/11
to unma...@googlegroups.com
Dear Jorge,

not as of now, as far as I know.

There are two models for species co-occurrence (meaning: potential interaction), one of which is implemented in PRESENCE (not sure about MARK) and described in a paper by MacKenzie et al. (J Animal Ecology, 2004), while the other can easily be implemented in WinBUGS (and probably also in a likelihood framework). The latter model is described by Waddle et al., Ecol. Appl. 2010.

Of course, there is also the multi-species site-occ framework by Dorazio and Royle, but I presume this is not what you mean.

Kind regards --- MArc


______________________________________________________________

Marc Kéry
Tel. ++41 41 462 97 93
marc...@vogelwarte.ch
www.vogelwarte.ch

Swiss Ornithological Institute | Seerose 1 | CH-6204 Sempach | Switzerland
______________________________________________________________

***** Check out my recent book on Bayesian statistical modeling at www.mbr-pwrc.usgs.gov/pubanalysis/kerybook

***** Watch out for our new book (due Dec 2011): Kéry & Schaub, Bayesian Population Analysis using WinBUGS, Academic Press;
www.vogelwarte.ch/bpa

________________________________________
From: unma...@googlegroups.com [unma...@googlegroups.com] on behalf of Jorge A. Ahumada [j.ah...@conservation.org]
Sent: 19 September 2011 21:36
To: Unmarked package
Subject: [unmarked] Species co-occurrence models

Matthew Giovanni

unread,
Sep 19, 2011, 3:56:56 PM9/19/11
to unma...@googlegroups.com
The way Jorge states the question sort of makes it sound as simple as using abundance or occurrence data for one species as a covariate for modeling like parameters for another species, but that's just my knuckle-dragger impulse kicking in.  There is also Zipkin et al.'s (2009 Journal of Applied Ecology) model of community or multi-species occurrence, which might be useful depending on your question.

Matt

Jeffrey Royle

unread,
Sep 19, 2011, 4:08:57 PM9/19/11
to unma...@googlegroups.com
unmarked does not do this stuff yet.
this would be really easy to implement if someone has time to get involved!
otherwise it will be awhile before we get to it...
andy

Andy Royle

unread,
Sep 19, 2011, 6:55:35 PM9/19/11
to unma...@googlegroups.com

hi Matt,
 you're essentially correct, at least conceptually: The Waddle and Dorazio models do use "occurrence of species B" as a covariate for occurrence of species A.
 the key technical issue is that you cannot observe occurrence of species B perfectly , so the likelihood has to be expressed as a mixture of potential species B occurrence states. (or analyzed conditional on occurrence of B using MCMC).
regards,
andy

 
-----unma...@googlegroups.com wrote: -----

To: unma...@googlegroups.com
From: Matthew Giovanni <matthew...@gmail.com>
Sent by: unma...@googlegroups.com
Date: 09/19/2011 03:56PM
Subject: Re: [unmarked] RE: Species co-occurrence models
Reply all
Reply to author
Forward
0 new messages