sam.garvin wrote:Thank you for your response! In my study, I am assuming that there is no significant difference between the types of sites ie. urban/rural, but I do want to account for the distance between sites. So is there no way to account for this proximity of sites in a single season model? If sites are neighbors, then the detection probability for the species may be over-estimated as a whole.
Sam,
If your looking for a method that incorporates distance between sites into the occupancy estimation procedure, I worked up some WinBUGS code that relies on penalized splines which works for a single season occupancy model that treats the matrix of distances between and among all sites and a mesh of fixed locations as a random effects term for space. I used the spatial process on the occurrence predictions, but using it on the detection prediction would be fairly straightforward.
The paper is: Collier et al. 2012: Diversity and Distributions 18: 158-167;
you can download the WinBUGS code example from
http://irnr.tamu.edu/bret/BretWebSiteDocs/WinBUGSCode.txtbret