www.phidot.org

Hello Murray!

I am interested in identifying nest sites locations of bumblebees using worker capture recapture data, and though that I might be able to use Secr models to identify the center of home range for workers as a proxy for location of next site. I was exploring the home range statistics section of the Secr package and was wondering if it would be possible to calculate the coordinates of home range centers for individuals(aka a single hive) in a population using this package? I have reviewed the literature and have not observed similar use of Secr.

Thanks,
Genevieve

Hi
This is a good use for the 'fxi' functions in 'secr' - basically, given a fitted model for the detection process you can infer the most likely locations of the unobserved home range centres of each animal you detected (but not, of course, of the ones you didn't detect). This is better than just taking the centroid of the observed locations as it allows for incursions from animals centred off-grid, and provides something like an asymmetric spatial confidence interval (contours of probability conditional on the model).
Murray

Example for first 5 animals in demo dataset:

Code: Select all

fxi.output <- fxi.contour(secrdemo.0, 1:5, fitmode = TRUE)
t(sapply(fxi.output, '[[', 'mode'))
     x        y       
[1,] 318.9968 609.0508
[2,] 618.1745 619.0023
[3,] 600.1212 515.6552
[4,] 499.9998 499.9998
[5,] 454.9836 614.2455