Table 1 Top models (ΔAICc < 2) of detection (p), occupancy per km (ψ) and abundance per km (λ) for grass snakes in Jersey. Parameters (p, ψ and λ) were constant (.) or allowed to vary with covariates. Due to the small sample size, models are ranked by their AICc and weight (w _i ). Models are shown for number of observations set to 132 (total number of surveys) for detection and 19 (number of sites) for occupancy and abundance. All models include an offset for transect length on occupancy or abundance. N is number of parameters in the model and LL the log-likelihood. The mean prediction and its standard error are shown for each parameter. Goodness-of-fit statistics are also shown.

From: Optimising monitoring efforts for secretive snakes: a comparison of occupancy and N-mixture models for assessment of population status