Occupancy Models

In this chapter, we will introduce occupancy models (MacKenzie et al. 2002; Tyre et al. 2003) and extensions that use detection/non-detection data to estimate the true occupancy status of species. Occupancy is probably the simplest way to measure population status and thus provides limited information for the spatial structures and dynamics of populations. Occupancy status, which is often represented by 1’s (presence) and 0’s (absence), is a reduced form of abundance in the sense that the study species will be considered present at a site as long as there is at least one individual of the species at that site (i.e., local abundance ≥1), and absent only when there is no individual at all at that site (i.e., local abundance = 0). Dynamic occupancy models (MacKenzie et al. 2003; Royle and Kéry 2007) further explain the temporal changes of occupancy status with local extinction (change from 1 to 0) and colonization (change from 0 to 1), which again is a reduced form of the BIDE processes in the sense that a local population goes extinct when all individuals at that site either die (D) or emigrate (E) without giving birth (B) to any new individual, and an unoccupied site is colonized when at least one individual immigrates (I) into this site. Despite the limited information occupancy models can provide, they are nonetheless one of the most broadly used family of models in population ecology, due to the relative ease with which detection/non-detection data can be collected.