Portal de Periódicos da CAPES

Correcting observer effect bias in estimates of nesting success of a coastal bird, the White-fronted Plover Charadrius marginatus

2002; Taylor & Francis; Volume: 49; Issue: 2 Linguagem: Inglês

10.1080/00063650209461256

1944-6705

Penn Lloyd, Éva E. Plagányi,

Species Distribution and Climate Change

Capsule A recently developed observer-effects model gives better estimates than the survival model. Aims To compare the performance of the traditional survival model with that of an observer-effects model in estimating nesting success. Methods We used survival data from 654 nests in a covariate-dependent extension of a maximum-likelihood estimator model to simultaneously estimate daily survival rate and observer effect on nesting success in four different sectors of a study site. Standard likelihood-ratio tests and Akaike's information criterion differences were used to compare model performances. Results A significant (negative) observer effect was detected in one of four sectors of the study area, and was probably attributable to small mammal predators. Despite significantly higher predation risk during the laying period, there was a significant positive observer effect during the laying period, contrasting with a weak negative effect during the incubation period, suggesting that an observer's tracks around unattended nests reduced predation risk. Likelihood-ratio tests indicated a significantly better fit using the observer effects model versus the Mayfield survival model in four of six scenarios. Conclusion The observer-effects model accounts for bias in estimates of nest survival rates, and is preferred over the survival model because the inclusion of observer effect as a variable does not over-parameterize the model.