Analysis of each site without covariates

questions concerning analysis/theory using program PRESENCE

Analysis of each site without covariates

Postby marioreyesjr » Mon May 04, 2020 1:13 am

I'm completely new with this and i'm working on a college investigation by myself: "Quetzal occupancy (Pharomacrus moccino de la Llave, 1832) in the Honduran northeast: automated bioacoustics monitoring"

i'm trying to use quetzal's territorial song to asses the presence/absence of this species within 16 days recording continuously. i'm willing to evaluated the Occupancy of each site (1, 2, 3, 4, 5 and 6), i don't know the sites so i can't use any sampling covariate nor site covariate (maybe altitude of each recorder,one of them below 1499 m.a.s.l. and others above 1500 m.a.s.l. but i don't know how).

It is anyway possible to get the Occupancy estimate of each site without using covariates? i need this information at least to make a inference on the quetzal occupancy.

This is my Presence/Absence Data:

(Site 1) 1 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1
(Site 2) 0 1 1 0 1 1 1 1 1 1 0 0 1 1 1 1
(Site 3) - - 0 1 0 1 0 0 0 1 0 1 0 0 0 0
(Site 4) - - 0 1 1 1 1 1 1 1 1 1 1 1 1 1
(Site 5) - - 1 1 1 1 1 1 1 1 1 1 1 1 1 1
(Site 6) 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1

I practiced with the GIANT WETA excercise (i finish it satisfactory), but i don't feel sure how i introduced the altitude as site covariate and i tried without positive results.

I will really appreciated your time and help! Thanks a lot
marioreyesjr
 
Posts: 4
Joined: Fri Apr 24, 2020 2:30 am

Re: Analysis of each site without covariates

Postby darryl » Mon May 04, 2020 1:33 am

Quetzal's were detected at all of your 6 sites, so occupancy is 100%. There's no variation to explain using covariates.
Cheers
Darryl
darryl
 
Posts: 495
Joined: Thu Jun 12, 2003 3:04 pm
Location: Dunedin, New Zealand

Re: Analysis of each site without covariates

Postby marioreyesjr » Mon May 04, 2020 2:09 am

Hi Mr. Darryl, i got you, but what about the psi (probability the species is present at a sampling unit)? can i still get it individually for each site using PRESENCE without covariates?

until now, running a simple single season i only get this:

Individual Site estimates of <psi>
Site estimate Std.err 95% conf. interval
psi 1 site1 : 1.0000 0.0000 0.0000 - 1.0000

============================================================

Individual Site estimates of <P[1]>
Site estimate Std.err 95% conf. interval
P[1] 1 site1 : 0.7889 0.0430 0.6925 - 0.8611
P[2] 1 site1 : 0.7889 0.0430 0.6925 - 0.8611

It says site 1 get PSI: 1, when i had 3 absences on that site or it is the 0.7889 the PSI for site 1?

i need the 6 sites PSI...

I really appreciate your help!

Thanks
marioreyesjr
 
Posts: 4
Joined: Fri Apr 24, 2020 2:30 am

Re: Analysis of each site without covariates

Postby jhines » Mon May 04, 2020 7:59 am

With the single-season models in Presence, we assume that the species is either present or absent at a site for the entire survey period. Since the species was present and detected at site 1 on at least 1 survey, it must have been present at the site for all surveys. The 3 surveys with 0's are surveys where the species was present, but not detected (i.e., either none of the individuals made sounds near the recorders or the recorders malfunctioned).

It's not possible to estimate occupancy at a site at a point in time without some form of replication. The multi-method model uses multiple detection methods to give an estimate of local occupancy at a site at each survey. This model gives an estimate of 'psi' (probability that the species is present at a site), 'theta' (probability that the species is available to be detected (making noise in your case) at a site during a survey and 'p' (probability that the recorders picked up the sound) at the site during the survey.

With any of these models, you still need variation in the data in order to get an estimate of a parameter which varies by some covariate. If there is no variation in the data (nearly all 1's or all 0's), then there isn't enough variation to estimate something as a function of any covariate, only estimate as a constant.

Jim
jhines
 
Posts: 599
Joined: Fri May 16, 2003 9:24 am
Location: Laurel, MD, USA


Return to analysis help

Who is online

Users browsing this forum: No registered users and 11 guests