Data entry for Single season two species model

questions concerning analysis/theory using program PRESENCE

Data entry for Single season two species model

Postby jasniyas » Sat Feb 25, 2017 12:45 am

Hi all,

I am currently learning how to use the PRESENCE software. Therefore, pardon my question if very basic. I am trying to set up a model to analyze the co-occurrence rate of two species. I have the presence/absence data with a sampling occasion of 24 hrs for each of the species from 250 camera stations and each of these stations has been active for about 6-8 weeks. However, these cameras were not all active at the same time i.e. they were all activated on different days even months apart. My question is with regards to entering the data in the Presence input file. Should I take the dates into consideration while entering each value? Since some cameras might be active while the others are not.
Also, should I enter a "0" for each sampling occasion the camera is inactive?

Thank you.
Jasniya
jasniyas
 
Posts: 2
Joined: Wed Feb 22, 2017 7:25 pm

Re: Data entry for Single season two species model

Postby jhines » Sun Feb 26, 2017 1:01 pm

Hi Jasniya,

First, and most importantly, you don't want to enter '0' when you did not visit a site. That will create a negative bias in detection probability and positive bias in estimated occupancy. PRESENCE accepts a '-' or '.' in the detection history for a "missing" value (ie. species presence not assessed during a survey).

If you're binning each 24-hour interval into surveys, then you could build detection history records with the missing value character for each day the camera was not operating at each site. This seems to be the most logical approach. With 6-8 weeks of "trapping", you might want to bin into "weeks" instead of days so you don't have so many zeros and/or missing values in the data.

Another possibility is to ignore calendar day and make survey 1 = 1st day of sampling for each site. If you only sampled a few weeks at each site, but the sampling was staggered among sites, this would reduce the number of surveys, but you would lose the ability to model date-specific detection (since survey1 would not be the same day for all sites).
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 9 guests