A Demo-Genetic Analysis of a Small Reintroduced Carnivore Population: The Otter (Lutra lutra) in The Netherlands

Table 1

Model selection (two-stage models) for survival (𝑆), resighting probability (𝑝), and recovery probability (𝑟) with the site fidelity (𝐹) of the otters set to 1. For each model the corrected quasi likelihood Akaike’s information criterion (QAICc), the difference in QAICc between the current model and the best fitting model (ΔQAICc), the number of estimable parameters (np), their weight (QIACc weight), and deviance (Qdev) are given. Fitted models are sorted by increasing QAICc, the first being the one that best represents the data. We present here only those models for which QAICc weight was greater than 0.01. These results are obtained after adjusting for overdispersion with 𝑐̂=1.338.

Formula	np	QAICc	ΔQAICc	QAICc weight	QDev
S(~sex * stage) p(~stage) r(~1)	7	524.367	0	0.478	386.088
S(~sex * stage) p(~stage) r(~sex)	8	525.607	1.240	0.257	385.191
S(~sex * stage) p(~sex * stage) r(~1)	9	527.074	2.706	0.123	384.504
** S(~sex * stage) p(~sex * stage) r(~sex)	10	528.236	3.869	0.069	383.495
S(~sex) p(~stage) r(~1)	5	531.709	7.342	0.012	397.648
S(~stage) p(~stage) r(~1)	5	532.056	7.689	0.010	397.995

**denotes the most general model and bold characters the two models used for estimating the parameter values (see Table 2).