Bias adjustment in Bayesian estimation of bird nest age-specific survival rates.

The populations of many North American landbirds are showing signs of declining. Gathering information on breeding productivity allows early detection of unhealthy populations and helps develop good habitat-management practices. In this paper, we study the performance of the Bayesian model (He, 2003, Biometrics 59, 962-973) for age-specific nest survival rates with irregular visits. We find that the estimates are satisfactory except for the age-one survival rate. Usually the more days skipped between two visits, the more serious the underestimation of the age-one survival rate. We investigated the problem and developed three approaches to adjust for the underestimation bias. The simulation results show that the three approaches can significantly improve the estimation of the age-one survival rate.