occupancy-and-detection

To estimate site occupancy and detection probability from repeat surveys, you fit occupancy models that account for imperfect detection. This process uses detection history matrices and site covariates to produce psi and p estimates.

Imperfect detection in occupancy modeling means a species is present at a site but missed during a survey. Accounting for it prevents underestimating true site occupancy rates by separating the probability of presence from the probability of detection.

You can run dynamic occupancy models to estimate colonization and extinction rates using Python interfaces to occupancy modeling tools. These models analyze multi-season detection histories alongside covariates to estimate changing occupancy dynamics.

To include site covariates in single-season occupancy models, you specify them for both occupancy and detection parameters. This allows the model to estimate how environmental variables influence site occupancy probability and survey-specific detection probability.

Goodness-of-fit testing and model selection for occupancy models involve generating c-hat diagnostics and model selection tables. These outputs assess model adequacy and compare competing models using covariate effects and detection estimates.

You can use the unmarked package or Python interfaces to analyze detection histories for psi estimation. These tools provide the necessary functions to process repeat survey data, fit models, and output occupancy estimates.