Royle point-count survey for territorial birds

questions concerning analysis/theory using program PRESENCE

Re: Royle point-count survey for territorial birds

Postby atsw2 » Fri Jan 20, 2012 4:57 pm

Hi,

I was trying to use the other models of Presence and I was wondering... could I use the False-Positive detections model? I could use the individual site estimate (psi) with the site area and the island area to calculate the abundance of this specie on the whole island. We calculated the island area (4698 ha) and a site would be 6 ha, it's just the individual estimate of a site which is a problem since our data is too sparse... I mean, with lambda=3.3, it means that the island has 2500 birds, which seems improbable (in 2003, they estimated 600 birds...)

Thank you very much for you're help
atsw2
 
Posts: 6
Joined: Mon Jan 02, 2012 4:34 pm

Re: Royle point-count survey for territorial birds

Postby darryl » Sun Jan 22, 2012 5:04 pm

Do you have certain and uncertain detections? If the model assumptions are reasonable for your situation then possibly yes, but the data you posted earlier was certainly pretty sparse already so your confidence intervals are likely to be very wide.

You asked an earlier question about the detection function; that goes hand-in-hand with the estimate of lambda, so if one is unreasonable/inaccurate, the other is likely to be too.

Cheers
Darryl
darryl
 
Posts: 498
Joined: Thu Jun 12, 2003 3:04 pm
Location: Dunedin, New Zealand

Re: Royle point-count survey for territorial birds

Postby atsw2 » Mon Jan 23, 2012 6:58 pm

Hi,

Yes some of our data had to be adapted because they could not be visited more than once (since these areas were only accessible by boat). We stayed one hour at these areas, and adapted them into 3 visits, so these can be counted as uncertain data. We have some data outside point-counts, and are uncertain data too. I did not add them before into the previous data because there were outside the point-counts and so it was not sure they occupied that territory or not.

With the data that was added, I've run the False-Positive model, and the confidence interval is less spread out. I've calculated with the individual estimate and got 443 birds [393;476] (for 9ha, instead of 6ha). I think this is a good estimation of the population. But we will definitely need to increase the number of surveys next year or change our detection method because it does not seem to be adapted to this specie.

Thank you so much for you're help!
atsw2
 
Posts: 6
Joined: Mon Jan 02, 2012 4:34 pm

Previous

Return to analysis help

Who is online

Users browsing this forum: No registered users and 2 guests