Objectives

In this unit we will use the PRESENCE software package to model present/absence data, extracting the detection probability parameter and getting the occupancy rate for various habitat types.

PRESENCE uses maximum likelihood estimators to obtain the best estimates of the model parameters, then compares the models using AIC. You should be familiar with these concepts already: check the units "Frogs in ponds - maximum likelihood estimators" and "More frogs in ponds - AIC and likelihood" if necessary.

Working through the analysis

From December 1997 to October 1999, a major camera trapping survey was carried out at 9 sites in Peninsular Malaysia (Lynam et al 2007). The main purpose of the survey was to ascertain the status of tigers, but of course the camera traps picked up many other species. In this unit we look at the camera trap results for golden cats.

Download the file "Golden_cats_PRESENCE.xls".

The "raw data" spreadsheet gives the results of camera trapping, with habitat information and a series of 1's and 0's indicating whether a golden cat was photographed on each day that the camera was out.

Download the lab guide "Golden_cats_PRESENCE.pdf". You will probably want to print out the lab guide and have it next to your computer while you work through the instructions.

Important! A new version of PRESENCE, version 2.1, was released on 16 Jan 2008. This has several improvements, including a much better method of calculating the chi-square statistic for assessing model fit. If you have version 2 installed (check Help > About), you should definitely upgrade. The lab guide has been updated to work with the new version, so you may need to download that again too.

Work through the lab guide before checking the results below.

Results

The best model according to the AIC criterion used the same parameter for p and psi for all forested sites. However, the models using different values of p or psi for primary vs. logged forest are also plausible, and model results can be averaged. There is little difference in occupancy between primary and logged forest. Detailed results are here.

The simulation shows that the number of camera traps deployed in plantations in this study is inadequate to sample golden cats, even assuming that occupancy in plantations is as high as in forested areas.

What next?

The first paper on these methods was MacKenzie et al (2002) and this is a good introduction. The book on the subject is MacKenzie et al (2006), which is a mine of information.

The PRESENCE software itself comes with several data sets and tutorials, based on the book: check the Help menu in PRESENCE.