RPresence

[Jocelyn Akins]

Hi folks,

Is anyone using Darryl Mackenzie's new R package RPResence for occupancy modeling? I am working through it now with my camera and trail transect data. Would you like to start an informal conversation thread and share our code? Let me know.

Jocelyn

[Jim Baldwin]

How do I download it?  I didn't see it in the CRAN repository.  (And my feeble googling didn't get anything to show.)

Jim

--
You received this message because you are subscribed to the Google Groups "unmarked" group.
To unsubscribe from this group and stop receiving emails from it, send an email to unmarked+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Gavin M. Jones]

I'm with Jim - I was unable to locate this package. Apparently some folks have used it - there is at least one reference to 'RPRESENCE by Mackenzie and Hines 2014' out on the web. Also on the PRESENCE software page (http://www.mbr-pwrc.usgs.gov/software/presence.html) there is a note that the most recent update (v11.0) better accommodates 'RPRESENCE'.  

Does anyone have more information?

[Brent P]

Any update on this? I know that he is using this package at his short-courses but I am unable to install...

[Mazerolle, Marc J.]

Brent,

see http://www.mbr-pwrc.usgs.gov/software/presence.shtml near bottom of the page (Linux and Windows versions of the package available).

Marc

--
____________________________________
Marc J. Mazerolle
Département des sciences du bois et de la forêt
2405 rue de la Terrasse
Université Laval
Québec, Québec G1V 0A6, Canada
Tel: (418) 656-2131 ext. 7120
Email: marc.ma...@uqat.ca

---

De : unma...@googlegroups.com [unma...@googlegroups.com] de la part de Brent P [bpe...@siu.edu]
Envoyé : vendredi 14 octobre 2016 14:29
À : unmarked
Objet : Re: [unmarked] RPresence

[Brent P]

Thank you, Marc!

[Jocelyn Akins]

Hi, I have been using RPresence and am happy to share thoughts and trouble shoot.

Jocelyn

[alex....@gmail.com]

Unmarked can run occupancy modeling as well.  Is the main analytical difference with Presence the ability to include random effects?  Are there other advantages/disadvantages to modeling occupancy in Presence vs Unmarked?  Thanks all!

-Alex

[Kery Marc]

Dear all,

I admit that I haven't yet checked out PResence.

However, it seems to be the R version of a Windows program called PRESENCE, which was created originally by Darryl MacKenzie and over the years now mostly by Jim Hines. However, I think that it's fair to say that, yes, unmarked can do quite a bit of occupancy modeling, but PRESENCE has many more occupancy models than does unmarked. For instance, multi-state occupancy models for static and dynamic situations, more false-positive occupancy modeling (I think, dynamic, too) and some slightly spatial models (e.g., the tiger on trails-model of Hines et al. 2010 and I think also a very simple version of an autologistic model, as was used in a recent publication by Yackulik et al). And no doubt quite some more. So I think it's a great thing that one now seems to be able to harness the power of PRESENCE from R, without having to click and point until your fingers hurt.

In terms of the models covered, the main thing where unmarked is 'better' than PRESENCE is in the models for abundance.

Apologies to Jim Hines, Richard Chandler and Ian Fiske for any mis-representation of what both are great pieces of software !

Best regards  --- Marc

---

From: unma...@googlegroups.com [unma...@googlegroups.com] on behalf of alex....@gmail.com [alex....@gmail.com]
Sent: 17 November 2016 17:18
To: unmarked
Subject: Re: [unmarked] RPresence

[alex....@gmail.com]

Hi Marc,

Thanks for your reply!  Hope you're doing well!

Best,

Alex

[Christos Astaras]

Hello Mark,

As far as I am aware at least, RPresence can't run the Hines et al. 2010 models on spatial-dependency of detections (suitable for spatial replicate surveys i.e. undertaken along paths/transects). Presence can, but last time I checked it was not possible in the R package. To me, that would have been by far the biggest asset for using the package but it is not there yet. I would love to hear from someone else that I am wrong as we use those models often (or at least consider them) at our unit.

Regards,

Christos

[Kery Marc]

Dear Christos,

thanks for the correction. In my mail below I was just describing some of the capabilities of the Windows version of PRESENCE. Hopefully, soon one may run all the models implemented in PRESENCE also in the R version.

Personally, I think the tigers on trails-model is fine, but my personal favourite of what one can do in PRESENCE but not in unmarked would be multistate occupancy models :)

Best regards  -- Marc

---

From: unma...@googlegroups.com [unma...@googlegroups.com] on behalf of Christos Astaras [christos...@gmail.com]
Sent: 18 November 2016 15:54

[Mathias Tobler]

All of these models (multi-state, multi-scale, correlated detections (Hines et al. 2010), robust design) are implemented in Mark and can be run through RMark. They have been available for a while now. RMark is a great package to use Mark through R without having to click through lots of menus and dialogs and the help file comes with many examples including one for multi-state occupancy models.

Best,

Mathias

[Kery Marc]

Hoi Mathias,

well, never forget MARK ! Do you know whether there is some repository of cookbook examples for such analyses with RMark ?

Best regards  ---- Marc

---

From: unma...@googlegroups.com [unma...@googlegroups.com] on behalf of Mathias Tobler [mato...@gmail.com]
Sent: 18 November 2016 18:08
To: unma...@googlegroups.com
Subject: Re: [unmarked] RPresence

[Mathias Tobler]

Hallo Mark,

there is Appendix C of the Mark book that has annotated examples including the multi-state occupancy model.

http://www.phidot.org/software/mark/docs/book/pdf/app_3.pdf

and the package help file has quite a few examples with detailed code and datasets for all occupancy models. e.g.

https://www.rdocumentation.org/packages/RMark/versions/2.2.0/topics/NicholsMSOccupancy

Best,

Mathias

On 18.11.2016 10:39, Kery Marc wrote:

[leara...@gmail.com]

Hi Jocelyn,

I like the idea of having a specific Rpresence group as I think that more people will start using it once they know it exists. 

Lea

[Evan Cooch]

The main support for RPresence (when Jim and Darryl are ready to 'unleash it' in wide release - that hasn't happened yet), will be on the PhiDot forum (same as for Presence, MARK, DENSITY, secr, and so on). 

If you don't know about the PhiiDot forum:  http://www.phidot.org/forum/index.php

[Evan Cooch]

Not as of yet. Larissa Bailey and students are working on the 'occupancy' chapter for the MARK book -- it will have lots of examples (in the fashion of your books, and the MARK book in general). Interfacing with RMark is relatively straightforward, but I suspect jeff Laake will cobble together some RMark-specific text to accompany the occupancy material.

On Friday, November 18, 2016 at 1:40:01 PM UTC-5, Kery Marc wrote:

Hoi Mathias,

well, never forget MARK ! Do you know whether there is some repository of cookbook examples for such analyses with RMark ?

Best regards  ---- Marc

---

From: unma...@googlegroups.com [unma...@googlegroups.com] on behalf of Mathias Tobler [mato...@gmail.com]
Sent: 18 November 2016 18:08
To: unma...@googlegroups.com
Subject: Re: [unmarked] RPresence

All of these models (multi-state, multi-scale, correlated detections (Hines et al. 2010), robust design) are implemented in Mark and can be run through RMark. They have been available for a while now. RMark is a great package to use Mark through R without having to click through lots of menus and dialogs and the help file comes with many examples including one for multi-state occupancy models.

Best,

Mathias




On 18.11.2016 07:16, Kery Marc wrote:

Dear Christos,

thanks for the correction. In my mail below I was just describing some of the capabilities of the Windows version of PRESENCE. Hopefully, soon one may run all the models implemented in PRESENCE also in the R version.

Personally, I think the tigers on trails-model is fine, but my personal favourite of what one can do in PRESENCE but not in unmarked would be multistate occupancy models :)

Best regards  -- Marc

---

From: unm...@googlegroups.com [unma...@googlegroups.com] on behalf of Christos Astaras [christos...@gmail.com]

[Kery Marc]

Dear Evan,

thanks for the important additional information. Since MARK is powerful on occupancy models as well, it will be great for many users to run them from R.

Best regards  ---- Marc

---

From: unma...@googlegroups.com [unma...@googlegroups.com] on behalf of Evan Cooch [evan....@gmail.com]
Sent: 06 January 2017 02:53

[Jennie Miller]

Hi everyone,

I'm using RPresence to run single-species and 2-species single-season models, and having trouble figuring out how to format and model "seasonal" site covariates for occupancy, or covariates that change across the surveys.

I've seen a few posts* that addressed this topic but can't figure out how to code it in RPresence. In one post, I saw that Jim suggested coding the seasonal site covariates as sampling covariates, but I get an error in RPresence when I input the site covariates in the sampling covariate (sampling covariates are a long list, e.g. 1 column for the length of sites/stations * n surveys, whereas site covariates are input as a frame consisting of a single column for each covariate for the length of sites).

Has anyone modeled seasonal site covariates with success in RPresence? I'm happy to share more specific code to troubleshoot the problem but since this is a general data formatting question it should be consistent across examples.

Thanks for starting this group - will be very helpful until there's greater documentation and help available for the product once it's formally released.

Thanks,
Jennie

*Related posts for PRESENCE:
http://www.phidot.org/forum/viewtopic.php?f=11&t=1134&p=3054&hilit=jim+multi#p3054
http://www.phidot.org/forum/viewtopic.php?f=11&t=2117

[Evan Cooch]

This is *not* a support forum for RPresence. The official support site for that is the RPresence subforum on http://www.phidot.org/forum 

I would suggest you post your question there. Produce (being RPresence) is 'live in the wild', and Darryl uses it now in workshops (including the one he's holding this week here at Cornell), as (I suspect) does Jim Hines.