Chapter 6: Fitting models using secr
Part 1. Data setup
The R package secr provides methods for estimating animal density and detectability from SCR data undermany different conditions. First, you need to get your data into the format secr wants. There are 3 key input files
​
-
A "trapfile" containing camera trap locations and any covariates associated with them.
​
-
A "capthist" file containing records of animal encounters at cameras.
​
-
A "mask" (also sometimes called a "mesh" or "state space"), a discretised grid of points used for numerically maximizing the SCR likelihood, together with any covariates associated with these points.
​
A quick-reference "cheatsheet" to supplement this video can be down loaded from this page.
Part 2. Fitting models
We cover the basics of using package secr to estimate the parameters of the SCR model and draw inferences about animal density and detectability.
​
A quick-reference "cheatsheet" to supplement this video can be down loaded from this page.