Linking habitat suitability to demography in a pond-breeding amphibian

Table 2 Selection of multiseason-multistate models for estimating occupancy and reproduction probabilities of great crested newts

Model	AIC	∆AIC	w	K
ψ (HSI), R (HSI), δ_s, p ^[1] (CE), p ^[2] (CE)	261.34	0.00	0.77	12
ψ (.), R (HSI), δ_s, p ^[1] (CE), p ^[2] (CE)	263.89	2.55	0.22	11
ψ (HSI), R (.), δ_s, p ^[1] (CE), p ^[2] (CE)	269.31	7.97	0.02	11
ψ (.), R (.), δ_s, p ^[1] (CE), p ^[2] (CE)	278.04	16.70	0.00	10

Probability of pond occupancy (ψ) and probability of reproduction, given presence (R) were held constant (.) or modelled as functions of habitat suitability index (HSI). The structure of the top-tanking model for detection probabilities {ψ (.), R (.), δ_s, p ^[1] (CE), p ^[2] (CE)} was used to evaluate the effect of HSI on ψ and R. Probability of correctly identifying a site as breeding site, given successful reproduction (δ) was modelled different in each capture period and probabilities of detecting occupancy, given occupancy without reproduction (p ^[1]) and with successful reproduction (p ^[2]) were modelled as functions of the number of capture events per capture period (CE). AIC: Akaike’s information criterion; ∆AIC: difference of the AIC value of the current and the best model; w: AIC weight; K: number of parameters.